From 56feaf002b98cc50d604188343d3e618d6aea40a Mon Sep 17 00:00:00 2001
From: Alexandros Ladas <alexandros@getml.com>
Date: Fri, 6 Sep 2024 18:34:15 +0200
Subject: [PATCH 1/3] Fix deprecation warnings and update benchmark notebooks
 to getml 1.5.0

---
 fastprop_benchmark/air_pollution_prop.ipynb |  976 +++++++++++----
 fastprop_benchmark/dodgers_prop.ipynb       | 1021 ++++++++++++----
 fastprop_benchmark/interstate94_prop.ipynb  |  763 +++++++++---
 fastprop_benchmark/occupancy_prop.ipynb     | 1212 +++++++++++++++----
 fastprop_benchmark/robot_prop.ipynb         | 1192 +++++++++++++-----
 utils/ft_time_series_builder.py             |    1 +
 6 files changed, 3995 insertions(+), 1170 deletions(-)

diff --git a/fastprop_benchmark/air_pollution_prop.ipynb b/fastprop_benchmark/air_pollution_prop.ipynb
index 2fbadb8..b13be43 100644
--- a/fastprop_benchmark/air_pollution_prop.ipynb
+++ b/fastprop_benchmark/air_pollution_prop.ipynb
@@ -21,11 +21,24 @@
     "- Population size: __41757__"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": [
+     "remove_cell_on_docs"
+    ]
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/air_pollution_prop.ipynb\">\n",
+    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Background\n",
+    "## Background\n",
     "\n",
     "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
     "\n",
@@ -38,14 +51,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Analysis"
+    "## Analysis"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Table of contents\n",
+    "Table of contents\n",
     "\n",
     "1. [Loading data](#1.-Loading-data)\n",
     "2. [Predictive modeling](#2.-Predictive-modeling)\n",
@@ -56,39 +69,70 @@
   {
    "cell_type": "code",
    "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
    "metadata": {
     "tags": []
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML API version: 1.5.0\n",
+      "\n"
+     ]
+    }
+   ],
    "source": [
-    "import datetime\n",
     "import os\n",
+    "import sys\n",
+    "\n",
     "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
     "from pathlib import Path\n",
-    "\n",
-    "import sys\n",
-    "import time\n",
     "from urllib import request\n",
-    "\n",
     "import getml\n",
-    "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
     "import pandas as pd\n",
-    "from scipy.stats import pearsonr\n",
     "\n",
-    "%matplotlib inline"
+    "print(f\"getML API version: {getml.__version__}\\n\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
     "parent = Path(os.getcwd()).parent.as_posix()\n",
     "\n",
     "if parent not in sys.path:\n",
-    "   sys.path.append(parent) \n",
+    "    sys.path.append(parent)\n",
     "\n",
     "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
    ]
@@ -97,21 +141,21 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 1. Loading data"
+    "### 1. Loading data"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.1 Download from source\n",
+    "#### 1.1 Download from source\n",
     "\n",
     "We begin by downloading the data from the UCI Machine Learning repository."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -127,7 +171,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -142,7 +186,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -159,14 +203,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.2 Prepare data for tsfresh and getML\n",
+    "#### 1.2 Prepare data for tsfresh and getML\n",
     "\n",
     "Our our goal is to predict the pm2.5 concentration from factors such as weather or time of day. However, there are some **missing entries** for pm2.5, so we get rid of them."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -186,7 +230,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -227,7 +271,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -250,7 +294,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -261,7 +305,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -282,7 +326,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -472,7 +516,7 @@
        "[41757 rows x 9 columns]"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -490,28 +534,54 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML engine is already running.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'air_pollution'\n"
+      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-community-edition-linux...\n",
+      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240905115118.log.\n",
+      "\u001b[2K  Loading pipelines... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 83%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'air_pollution'</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Connected to project \u001b[32m'air_pollution'\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "getml.engine.launch(home_directory=Path.home(), allow_remote_ips=True, token='token')\n",
+    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
     "getml.engine.set_project(\"air_pollution\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -528,7 +598,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -2001,7 +2071,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2017,7 +2087,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
@@ -2154,7 +2224,7 @@
        "type: StringColumnView"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2170,19 +2240,19 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 2. Predictive modeling"
+    "### 2. Predictive modeling"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.1 Propositionalization with getML's FastProp"
+    "#### 2.1 Propositionalization with getML's FastProp"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
@@ -2318,7 +2388,7 @@
        "        \n",
        "      \n",
        "        \n",
-       "          <th class=\"int\"> rows</th>\n",
+       "          <th class=\"str\">rows   </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -2342,7 +2412,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">8661</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2363,7 +2433,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">33096</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2481,16 +2551,16 @@
        "container\n",
        "\n",
        "  population\n",
-       "      subset   name    rows   type\n",
-       "  0   test     full    8661   View\n",
-       "  1   train    full   33096   View\n",
+       "      subset   name   rows      type\n",
+       "  0   test     full   unknown   View\n",
+       "  1   train    full   unknown   View\n",
        "\n",
        "  peripheral\n",
        "      name    rows   type     \n",
        "  0   full   41757   DataFrame"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2509,7 +2579,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -2539,7 +2609,7 @@
        "         tags=['memory: 1d', 'simple features'])"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2562,19 +2632,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2583,7 +2690,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2592,28 +2699,115 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 331 features... 100% |██████████| [elapsed: 00:05, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:5.370066\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Trying 331 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:07\u001b[0m0%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:07.707196.\n",
       "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:03, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:07\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2624,18 +2818,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2644,7 +2851,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2657,56 +2864,130 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:16, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:16.267519\n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])</pre>"
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
       ],
       "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])"
+       "\n"
       ]
      },
-     "execution_count": 23,
      "metadata": {},
-     "output_type": "execute_result"
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:27\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:27.386765.\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre>Pipeline(data_model='population',\n",
+       "         feature_learners=[],\n",
+       "         feature_selectors=[],\n",
+       "         include_categorical=False,\n",
+       "         loss_function='SquareLoss',\n",
+       "         peripheral=[],\n",
+       "         predictors=['XGBoostRegressor'],\n",
+       "         preprocessors=[],\n",
+       "         share_selected_features=0.5,\n",
+       "         tags=['prediction', 'fastprop'])</pre>"
+      ],
+      "text/plain": [
+       "Pipeline(data_model='population',\n",
+       "         feature_learners=[],\n",
+       "         feature_selectors=[],\n",
+       "         include_categorical=False,\n",
+       "         loss_function='SquareLoss',\n",
+       "         peripheral=[],\n",
+       "         predictors=['XGBoostRegressor'],\n",
+       "         preprocessors=[],\n",
+       "         share_selected_features=0.5,\n",
+       "         tags=['prediction', 'fastprop'])"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
@@ -2715,18 +2996,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:27\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2796,7 +3090,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:14</td>\n",
+       "              <td class=\"datetime\">2024-09-05 11:52:05</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2825,7 +3119,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:14</td>\n",
+       "              <td class=\"datetime\">2024-09-05 11:52:06</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2855,11 +3149,11 @@
       ],
       "text/plain": [
        "    date time             set used         target        mae       rmse   rsquared\n",
-       "0   2024-02-21 16:13:14   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
-       "1   2024-02-21 16:13:14   fastprop_test    pm2.5     44.2526    63.4191     0.5462"
+       "0   2024-09-05 11:52:05   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
+       "1   2024-09-05 11:52:06   fastprop_test    pm2.5     44.2526    63.4191     0.5462"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2872,14 +3166,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.2 Using featuretools\n",
+    "#### 2.2 Using featuretools\n",
     "\n",
     "To make things a bit easier, we have written a high-level wrapper around featuretools which we placed in a separate module (`utils`)."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2895,7 +3189,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
@@ -2904,7 +3198,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 298 features...\n",
-      "Time taken: 0h:34m:1.838204\n",
+      "Time taken: 0h:33m:59.375927\n",
       "\n"
      ]
     }
@@ -2918,7 +3212,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2932,7 +3226,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2951,7 +3245,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -2981,7 +3275,7 @@
        "         tags=['featuretools', 'memory: 1d'])"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2998,19 +3292,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -3019,22 +3350,96 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:10, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:10.460135\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:09\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:09.898496.\n",
       "\n"
      ]
     },
@@ -3065,7 +3470,7 @@
        "         tags=['featuretools', 'memory: 1d'])"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3076,18 +3481,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:09\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -3157,7 +3575,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:56:07</td>\n",
+       "              <td class=\"datetime\">2024-09-05 12:34:12</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3186,7 +3604,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:56:07</td>\n",
+       "              <td class=\"datetime\">2024-09-05 12:34:12</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3216,11 +3634,11 @@
       ],
       "text/plain": [
        "    date time             set used                target        mae       rmse   rsquared\n",
-       "0   2024-02-21 16:56:07   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
-       "1   2024-02-21 16:56:07   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
+       "0   2024-09-05 12:34:12   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
+       "1   2024-09-05 12:34:12   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3233,7 +3651,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.3 Using tsfresh\n",
+    "#### 2.3 Using tsfresh\n",
     "\n",
     "tsfresh is a rather low-level library. To make things a bit easier, we have written a high-level wrapper which we placed in a separate module (`utils`).\n",
     "\n",
@@ -3251,7 +3669,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [
     {
@@ -3441,7 +3859,7 @@
        "[33096 rows x 9 columns]"
       ]
      },
-     "execution_count": 33,
+     "execution_count": 34,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3452,16 +3870,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:22<00:00,  1.82it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:26<00:00,  1.48it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:29<00:00,  1.35it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:20<00:00,  1.92it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.71it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.74it/s]\n"
      ]
     },
     {
@@ -3469,7 +3887,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 78 features...\n",
-      "Time taken: 0h:1m:30.572509\n",
+      "Time taken: 0h:1m:13.903742\n",
       "\n"
      ]
     },
@@ -3477,9 +3895,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.93it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.84it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:07<00:00,  5.61it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 16.30it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.04it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.02it/s]\n"
      ]
     }
    ],
@@ -3505,7 +3923,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3524,7 +3942,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3550,7 +3968,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
@@ -3580,7 +3998,7 @@
        "         tags=['tsfresh', 'memory: 1d'])"
       ]
      },
-     "execution_count": 37,
+     "execution_count": 38,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3597,19 +4015,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -3618,22 +4073,96 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:06, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:5.783919\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:05\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:05.505439.\n",
       "\n"
      ]
     },
@@ -3664,7 +4193,7 @@
        "         tags=['tsfresh', 'memory: 1d'])"
       ]
      },
-     "execution_count": 39,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3675,18 +4204,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:05\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -3756,7 +4298,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:58:06</td>\n",
+       "              <td class=\"datetime\">2024-09-05 12:35:50</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3785,7 +4327,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:58:06</td>\n",
+       "              <td class=\"datetime\">2024-09-05 12:35:50</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3815,11 +4357,11 @@
       ],
       "text/plain": [
        "    date time             set used           target        mae       rmse   rsquared\n",
-       "0   2024-02-21 16:58:06   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
-       "1   2024-02-21 16:58:06   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
+       "0   2024-09-05 12:35:50   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
+       "1   2024-09-05 12:35:50   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
       ]
      },
-     "execution_count": 40,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3832,12 +4374,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 3. Comparison"
+    "### 3. Comparison"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [
     {
@@ -3874,9 +4416,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:08.637513</td>\n",
+       "      <td>0 days 00:00:12.524776</td>\n",
        "      <td>289</td>\n",
-       "      <td>33.458244</td>\n",
+       "      <td>23.074438</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>44.252635</td>\n",
@@ -3885,22 +4427,22 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:34:01.839960</td>\n",
+       "      <td>0 days 00:33:59.377855</td>\n",
        "      <td>114</td>\n",
-       "      <td>0.055832</td>\n",
-       "      <td>236.392114</td>\n",
-       "      <td>599.266495</td>\n",
+       "      <td>0.055899</td>\n",
+       "      <td>162.827491</td>\n",
+       "      <td>412.785062</td>\n",
        "      <td>43.915071</td>\n",
        "      <td>62.567175</td>\n",
        "      <td>0.559369</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:01:30.572946</td>\n",
+       "      <td>0 days 00:01:13.904058</td>\n",
        "      <td>72</td>\n",
-       "      <td>0.794939</td>\n",
-       "      <td>10.485998</td>\n",
-       "      <td>42.089066</td>\n",
+       "      <td>0.974236</td>\n",
+       "      <td>5.900629</td>\n",
+       "      <td>23.684642</td>\n",
        "      <td>47.110594</td>\n",
        "      <td>66.599982</td>\n",
        "      <td>0.501524</td>\n",
@@ -3911,14 +4453,14 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:08.637513           289            33.458244   \n",
-       "featuretools    0 days 00:34:01.839960           114             0.055832   \n",
-       "tsfresh         0 days 00:01:30.572946            72             0.794939   \n",
+       "getML: FastProp 0 days 00:00:12.524776           289            23.074438   \n",
+       "featuretools    0 days 00:33:59.377855           114             0.055899   \n",
+       "tsfresh         0 days 00:01:13.904058            72             0.974236   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  \\\n",
        "getML: FastProp            1.000000                        1.000000   \n",
-       "featuretools             236.392114                      599.266495   \n",
-       "tsfresh                   10.485998                       42.089066   \n",
+       "featuretools             162.827491                      412.785062   \n",
+       "tsfresh                    5.900629                       23.684642   \n",
        "\n",
        "                       mae       rmse  rsquared  \n",
        "getML: FastProp  44.252635  63.419113  0.546164  \n",
@@ -3926,7 +4468,7 @@
        "tsfresh          47.110594  66.599982  0.501524  "
       ]
      },
-     "execution_count": 41,
+     "execution_count": 42,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3978,7 +4520,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 43,
    "metadata": {
     "tags": []
    },
@@ -3992,7 +4534,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 4. Conclusion\n",
+    "### 4. Conclusion\n",
     "\n",
     "We have compared getML's feature learning algorithms to tsfresh's brute-force feature engineering approaches on a data set related to air pollution in China. We found that getML significantly outperforms featuretools and tsfresh. These results are consistent with the view that feature learning can yield significant improvements over simple propositionalization approaches.\n",
     "\n",
@@ -4019,7 +4561,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.18"
+   "version": "3.11.9"
   }
  },
  "nbformat": 4,
diff --git a/fastprop_benchmark/dodgers_prop.ipynb b/fastprop_benchmark/dodgers_prop.ipynb
index adc3b82..b2cefb7 100644
--- a/fastprop_benchmark/dodgers_prop.ipynb
+++ b/fastprop_benchmark/dodgers_prop.ipynb
@@ -17,11 +17,24 @@
     "- Population size: __47497__"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": [
+     "remove_cell_on_docs"
+    ]
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/dodgers_prop.ipynb\">\n",
+    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Background\n",
+    "## Background\n",
     "\n",
     "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
     "\n",
@@ -34,15 +47,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Analysis"
+    "## Analysis"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Table of contents\n",
-    "\n",
     "1. [Loading data](#1.-Loading-data)\n",
     "2. [Predictive modeling](#2.-Predictive-modeling)\n",
     "3. [Comparison](#3.-Comparison)"
@@ -59,64 +70,114 @@
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML API version: 1.5.0\n",
+      "\n"
+     ]
+    }
+   ],
    "source": [
-    "import datetime\n",
-    "import gc\n",
     "import os\n",
+    "import sys\n",
+    "\n",
     "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
     "from pathlib import Path\n",
-    "\n",
-    "import sys\n",
-    "import time\n",
     "from urllib import request\n",
-    "\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "import numpy as np\n",
+    "import getml\n",
     "import pandas as pd\n",
-    "import scipy\n",
-    "from IPython.display import Image\n",
-    "from scipy.stats import pearsonr\n",
     "\n",
-    "%matplotlib inline"
+    "print(f\"getML API version: {getml.__version__}\\n\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
     "parent = Path(os.getcwd()).parent.as_posix()\n",
     "\n",
     "if parent not in sys.path:\n",
-    "   sys.path.append(parent) \n",
+    "    sys.path.append(parent)\n",
     "\n",
     "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/getml --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/getml/.getML/getml-1.4.0-x64-linux...\n",
-      "Launched the getML engine. The log output will be stored in /home/getml/.getML/logs/20240221161224.log.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'dodgers'\n"
+      "getML Engine is already running.\n",
+      "\u001b[2K  Loading pipelines... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 83%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'dodgers'</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Connected to project \u001b[32m'dodgers'\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "import getml\n",
-    "\n",
-    "getml.engine.launch(home_directory=Path.home(), allow_remote_ips=True, token='token')\n",
+    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
     "getml.engine.set_project(\"dodgers\")"
    ]
   },
@@ -124,21 +185,21 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 1. Loading data"
+    "### 1. Loading data"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.1 Download from source\n",
+    "#### 1.1 Download from source\n",
     "\n",
     "We begin by downloading the data from the UC Irvine Machine Learning repository:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -157,18 +218,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
-    "data_full_pandas[\"ds\"] = [\n",
-    "    datetime.datetime.strptime(dt, \"%m/%d/%Y %H:%M\") for dt in data_full_pandas[\"ds\"]\n",
-    "]"
+    "data_full_pandas[\"ds\"] = pd.to_datetime(data_full_pandas[\"ds\"], format=\"%m/%d/%Y %H:%M\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -274,7 +333,7 @@
        "[50400 rows x 2 columns]"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -292,7 +351,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -301,7 +360,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -311,7 +370,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -684,7 +743,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -695,7 +754,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -832,7 +891,7 @@
        "type: StringColumnView"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -848,7 +907,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.3 Define relational model\n",
+    "#### 1.3 Define relational model\n",
     "\n",
     "To start with relational learning, we need to specify the data model. We manually replicate the appropriate time series structure by setting time series related join conditions (`horizon`, `memory` and `allow_lagged_targets`). This is done abstractly using [Placeholders](https://docs.getml.com/latest/user_guide/data_model/data_model.html#placeholders)\n",
     "\n",
@@ -860,7 +919,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -996,7 +1055,7 @@
        "        \n",
        "      \n",
        "        \n",
-       "          <th class=\"int\"> rows</th>\n",
+       "          <th class=\"str\">rows   </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -1020,7 +1079,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">12384</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1041,7 +1100,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">38016</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1152,16 +1211,16 @@
        "container\n",
        "\n",
        "  population\n",
-       "      subset   name         rows   type\n",
-       "  0   test     data_full   12384   View\n",
-       "  1   train    data_full   38016   View\n",
+       "      subset   name        rows      type\n",
+       "  0   test     data_full   unknown   View\n",
+       "  1   train    data_full   unknown   View\n",
        "\n",
        "  peripheral\n",
        "      name         rows   type     \n",
        "  0   data_full   50400   DataFrame"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1190,7 +1249,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 2. Predictive modeling\n",
+    "### 2. Predictive modeling\n",
     "\n",
     "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
    ]
@@ -1199,12 +1258,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.1 Propositionalization with getML's FastProp"
+    "#### 2.1 Propositionalization with getML's FastProp"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1225,7 +1284,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [
     {
@@ -1255,7 +1314,7 @@
        "         tags=['feature learning', 'fastprop'])"
       ]
      },
-     "execution_count": 13,
+     "execution_count": 14,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1273,20 +1332,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1295,7 +1391,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1304,30 +1400,117 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 526 features... 100% |██████████| [elapsed: 00:06, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:6.317863\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Trying 526 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:08\u001b[0m5%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:08.361847.\n",
       "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:03, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:08\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1338,18 +1521,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1358,7 +1554,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1371,76 +1567,163 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:09, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:9.613381\n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])</pre>"
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
       ],
       "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])"
+       "\n"
       ]
      },
-     "execution_count": 19,
      "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.fit(fastprop_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:09\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:09.975733.\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre>Pipeline(data_model='population',\n",
+       "         feature_learners=[],\n",
+       "         feature_selectors=[],\n",
+       "         include_categorical=False,\n",
+       "         loss_function='SquareLoss',\n",
+       "         peripheral=[],\n",
+       "         predictors=['XGBoostRegressor'],\n",
+       "         preprocessors=[],\n",
+       "         share_selected_features=0.5,\n",
+       "         tags=['prediction', 'fastprop'])</pre>"
+      ],
+      "text/plain": [
+       "Pipeline(data_model='population',\n",
+       "         feature_learners=[],\n",
+       "         feature_selectors=[],\n",
+       "         include_categorical=False,\n",
+       "         loss_function='SquareLoss',\n",
+       "         peripheral=[],\n",
+       "         predictors=['XGBoostRegressor'],\n",
+       "         preprocessors=[],\n",
+       "         share_selected_features=0.5,\n",
+       "         tags=['prediction', 'fastprop'])"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pipe_fp_pr.fit(fastprop_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:09\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -1510,7 +1793,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:12:55</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:03:36</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1539,7 +1822,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:12:55</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:03:37</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1569,11 +1852,11 @@
       ],
       "text/plain": [
        "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:12:55   fastprop_train   y         5.4188    7.5347     0.699 \n",
-       "1   2024-02-21 16:12:55   fastprop_test    y         5.6151    7.8243     0.6747"
+       "0   2024-09-05 15:03:36   fastprop_train   y         5.4188    7.5347     0.699 \n",
+       "1   2024-09-05 15:03:37   fastprop_test    y         5.6151    7.8243     0.6747"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1586,12 +1869,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.2 Propositionalization with featuretools"
+    "#### 2.2 Propositionalization with featuretools"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1601,7 +1884,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1614,7 +1897,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1631,7 +1914,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -1640,7 +1923,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:9m:19.75259\n",
+      "Time taken: 0h:8m:58.96343\n",
       "\n"
      ]
     }
@@ -1654,7 +1937,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1669,7 +1952,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1684,7 +1967,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
@@ -1714,7 +1997,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1731,20 +2014,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -1823,7 +2143,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 28,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1834,24 +2154,122 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n",
-      "To see the issues in full, run .check() on the pipeline.\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:04, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:4.092266\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:03.767338.\n",
       "\n"
      ]
     },
@@ -1882,7 +2300,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1893,18 +2311,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -1974,7 +2405,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:25:31</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:15:36</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2003,7 +2434,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:25:31</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:15:36</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2033,11 +2464,11 @@
       ],
       "text/plain": [
        "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:25:31   featuretools_train   y         5.4482    7.568      0.6962\n",
-       "1   2024-02-21 16:25:31   featuretools_test    y         6.0863    8.5009     0.6498"
+       "0   2024-09-05 15:15:36   featuretools_train   y         5.4482    7.568      0.6962\n",
+       "1   2024-09-05 15:15:36   featuretools_test    y         6.0863    8.5009     0.6498"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2050,12 +2481,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.3 Propositionalization with tsfresh"
+    "#### 2.3 Propositionalization with tsfresh"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2072,16 +2503,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:19<00:00,  2.06it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:08<00:00,  4.69it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:08<00:00,  4.71it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:14<00:00,  2.83it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.78it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.93it/s]\n"
      ]
     },
     {
@@ -2089,7 +2520,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 13 features...\n",
-      "Time taken: 0h:0m:46.114942\n",
+      "Time taken: 0h:0m:34.625442\n",
       "\n"
      ]
     },
@@ -2097,9 +2528,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:05<00:00,  7.69it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 13.49it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:03<00:00, 12.04it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.28it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 18.13it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 17.05it/s]\n"
      ]
     }
    ],
@@ -2112,7 +2543,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2127,7 +2558,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2138,7 +2569,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [
     {
@@ -2168,7 +2599,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 35,
+     "execution_count": 36,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2183,25 +2614,130 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n",
-      "To see the issues in full, run .check() on the pipeline.\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:1.790984\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K\u001b[32m⠧\u001b[0m XGBoost: Trained tree 96.  \u001b[31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[35m 96%\u001b[0m • \u001b[36m00:01\u001b[0m"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:01.469758.\n",
       "\n"
      ]
     },
@@ -2232,7 +2768,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 37,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2243,18 +2779,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2324,7 +2873,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:26:34</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:16:24</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2353,7 +2902,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:26:34</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:16:24</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2383,11 +2932,11 @@
       ],
       "text/plain": [
        "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:26:34   tsfresh_train   y         6.3146    8.2348     0.6418\n",
-       "1   2024-02-21 16:26:34   tsfresh_test    y         6.7886    8.9134     0.5778"
+       "0   2024-09-05 15:16:24   tsfresh_train   y         6.3146    8.2348     0.6418\n",
+       "1   2024-09-05 15:16:24   tsfresh_test    y         6.7886    8.9134     0.5778"
       ]
      },
-     "execution_count": 37,
+     "execution_count": 38,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2400,12 +2949,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 3. Comparison"
+    "### 3. Comparison"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2453,7 +3002,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
     {
@@ -2490,9 +3039,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:09.406415</td>\n",
+       "      <td>0 days 00:00:11.918092</td>\n",
        "      <td>526</td>\n",
-       "      <td>55.919029</td>\n",
+       "      <td>44.134522</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.674740</td>\n",
@@ -2501,22 +3050,22 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:09:19.754041</td>\n",
+       "      <td>0 days 00:08:58.964287</td>\n",
        "      <td>59</td>\n",
-       "      <td>0.105403</td>\n",
-       "      <td>59.507691</td>\n",
-       "      <td>530.523794</td>\n",
+       "      <td>0.109469</td>\n",
+       "      <td>45.222363</td>\n",
+       "      <td>403.168329</td>\n",
        "      <td>0.649768</td>\n",
        "      <td>8.500887</td>\n",
        "      <td>6.086277</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:46.115063</td>\n",
+       "      <td>0 days 00:00:34.625580</td>\n",
        "      <td>12</td>\n",
-       "      <td>0.260219</td>\n",
-       "      <td>4.902512</td>\n",
-       "      <td>214.892468</td>\n",
+       "      <td>0.346565</td>\n",
+       "      <td>2.905296</td>\n",
+       "      <td>127.348619</td>\n",
        "      <td>0.577811</td>\n",
        "      <td>8.913408</td>\n",
        "      <td>6.788610</td>\n",
@@ -2527,14 +3076,14 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:09.406415           526            55.919029   \n",
-       "featuretools    0 days 00:09:19.754041            59             0.105403   \n",
-       "tsfresh         0 days 00:00:46.115063            12             0.260219   \n",
+       "getML: FastProp 0 days 00:00:11.918092           526            44.134522   \n",
+       "featuretools    0 days 00:08:58.964287            59             0.109469   \n",
+       "tsfresh         0 days 00:00:34.625580            12             0.346565   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.674740   \n",
-       "featuretools              59.507691                      530.523794  0.649768   \n",
-       "tsfresh                    4.902512                      214.892468  0.577811   \n",
+       "featuretools              45.222363                      403.168329  0.649768   \n",
+       "tsfresh                    2.905296                      127.348619  0.577811   \n",
        "\n",
        "                     rmse       mae  \n",
        "getML: FastProp  7.824273  5.615138  \n",
@@ -2542,7 +3091,7 @@
        "tsfresh          8.913408  6.788610  "
       ]
      },
-     "execution_count": 39,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2553,7 +3102,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2582,7 +3131,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.18"
+   "version": "3.11.9"
   },
   "toc": {
    "base_numbering": 1,
diff --git a/fastprop_benchmark/interstate94_prop.ipynb b/fastprop_benchmark/interstate94_prop.ipynb
index 3856403..7023907 100644
--- a/fastprop_benchmark/interstate94_prop.ipynb
+++ b/fastprop_benchmark/interstate94_prop.ipynb
@@ -17,11 +17,24 @@
     "- Population size: __24096__"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": [
+     "remove_cell_on_docs"
+    ]
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/interstate94_prop.ipynb\">\n",
+    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Background\n",
+    "## Background\n",
     "\n",
     "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
     "\n",
@@ -34,15 +47,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Analysis"
+    "## Analysis"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Table of contents\n",
-    "\n",
     "1. [Loading data](#1.-Loading-data)\n",
     "2. [Predictive modeling](#2.-Predictive-modeling)\n",
     "3. [Comparison](#3.-Comparison)"
@@ -64,83 +75,163 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML API version: 1.4.0\n",
-      "\n",
-      "getML engine is already running.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'interstate94'\n"
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML API version: 1.5.0\n",
+      "\n"
      ]
     }
    ],
    "source": [
-    "import datetime\n",
     "import os\n",
+    "import sys\n",
+    "\n",
     "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
     "from pathlib import Path\n",
     "\n",
-    "import sys\n",
-    "import time\n",
-    "\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "import numpy as np\n",
     "import pandas as pd\n",
-    "from IPython.display import Image\n",
-    "\n",
-    "%matplotlib inline\n",
-    "\n",
     "import getml\n",
     "\n",
-    "print(f\"getML API version: {getml.__version__}\\n\")\n",
-    "\n",
-    "getml.engine.launch(home_directory=Path.home(), allow_remote_ips=True, token='token')\n",
+    "print(f\"getML API version: {getml.__version__}\\n\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML Engine is already running.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'interstate94'</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Connected to project \u001b[32m'interstate94'\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
     "getml.engine.set_project(\"interstate94\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
     "parent = Path(os.getcwd()).parent.as_posix()\n",
     "\n",
     "if parent not in sys.path:\n",
-    "   sys.path.append(parent) \n",
+    "    sys.path.append(parent)\n",
     "\n",
-    "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder\n",
-    "import woodwork as ww"
+    "from utils import Benchmark, FTTimeSeriesBuilder"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 1. Loading data"
+    "### 1. Loading data"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.1 Download from source\n",
+    "#### 1.1 Download from source\n",
     "\n",
     "We begin by downloading the data from the UC Irvine Machine Learning repository:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading traffic<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Loading traffic\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\n",
-      "Loading traffic...\n",
-      " 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n"
+      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m1.2/1.2 MB\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[?25h"
      ]
     }
    ],
@@ -150,7 +241,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -159,7 +250,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -883,7 +974,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -896,12 +987,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.2 Define relational model\n"
+    "#### 1.2 Define relational model\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -910,7 +1001,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -1046,7 +1137,7 @@
        "        \n",
        "      \n",
        "        \n",
-       "          <th class=\"int\"> rows</th>\n",
+       "          <th class=\"str\">rows   </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -1070,7 +1161,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">4800</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1091,7 +1182,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">19296</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1210,16 +1301,16 @@
        "container\n",
        "\n",
        "  population\n",
-       "      subset   name       rows   type\n",
-       "  0   test     traffic    4800   View\n",
-       "  1   train    traffic   19296   View\n",
+       "      subset   name      rows      type\n",
+       "  0   test     traffic   unknown   View\n",
+       "  1   train    traffic   unknown   View\n",
        "\n",
        "  peripheral\n",
        "      name       rows   type     \n",
        "  0   traffic   24096   DataFrame"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1242,7 +1333,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 2. Predictive modeling\n",
+    "### 2. Predictive modeling\n",
     "\n",
     "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
    ]
@@ -1251,12 +1342,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.1 Propositionalization with getML's FastProp"
+    "#### 2.1 Propositionalization with getML's FastProp"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1277,7 +1368,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -1307,7 +1398,7 @@
        "         tags=['feature learning', 'fastprop'])"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1325,20 +1416,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1347,7 +1475,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1356,30 +1484,117 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 365 features... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:2.505792\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Trying 365 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0m2%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:03.025341.\n",
       "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1390,18 +1605,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1410,7 +1638,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1423,23 +1651,97 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:05, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:5.173095\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:05\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:05.318526.\n",
       "\n"
      ]
     },
@@ -1470,7 +1772,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1481,18 +1783,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:05\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -1562,7 +1877,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:15</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:56:19</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1591,7 +1906,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:15</td>\n",
+       "              <td class=\"datetime\">2024-09-05 15:56:19</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1603,11 +1918,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">180.4867</td>\n",
+       "              <td class=\"float\">180.5073</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">261.9389</td>\n",
+       "              <td class=\"float\">261.9896</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1621,11 +1936,11 @@
       ],
       "text/plain": [
        "    date time             set used         target                 mae        rmse   rsquared\n",
-       "0   2024-02-21 16:13:15   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
-       "1   2024-02-21 16:13:15   fastprop_test    traffic_volume    180.4867    261.9389     0.9827"
+       "0   2024-09-05 15:56:19   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
+       "1   2024-09-05 15:56:19   fastprop_test    traffic_volume    180.5073    261.9896     0.9827"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1638,12 +1953,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.2 Propositionalization with featuretools"
+    "#### 2.2 Propositionalization with featuretools"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1653,7 +1968,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1666,7 +1981,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1683,7 +1998,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 22,
    "metadata": {
     "tags": []
    },
@@ -1694,7 +2009,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:4m:46.444548\n",
+      "Time taken: 0h:4m:30.436553\n",
       "\n"
      ]
     }
@@ -1708,7 +2023,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1729,7 +2044,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1744,7 +2059,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -1774,7 +2089,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1791,19 +2106,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1812,22 +2164,96 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:2.260849\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:02.084733.\n",
       "\n"
      ]
     },
@@ -1858,7 +2284,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1869,18 +2295,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -1950,7 +2389,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:19:18</td>\n",
+       "              <td class=\"datetime\">2024-09-05 16:02:00</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1979,7 +2418,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:19:18</td>\n",
+       "              <td class=\"datetime\">2024-09-05 16:02:00</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2009,11 +2448,11 @@
       ],
       "text/plain": [
        "    date time             set used             target                 mae        rmse   rsquared\n",
-       "0   2024-02-21 16:19:18   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
-       "1   2024-02-21 16:19:18   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
+       "0   2024-09-05 16:02:00   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
+       "1   2024-09-05 16:02:00   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
       ]
      },
-     "execution_count": 26,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2026,7 +2465,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.3 Propositionalization with tsfresh\n",
+    "#### 2.3 Propositionalization with tsfresh\n",
     "\n",
     "tsfresh failed to run through due to an apparent bug in the tsfresh library and is therefore excluded from this analysis."
    ]
@@ -2035,12 +2474,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 3. Comparison"
+    "### 3. Comparison"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2081,7 +2520,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -2118,22 +2557,22 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:03.963151</td>\n",
+       "      <td>0 days 00:00:04.769646</td>\n",
        "      <td>461</td>\n",
-       "      <td>116.319646</td>\n",
+       "      <td>96.655712</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
-       "      <td>0.982678</td>\n",
-       "      <td>261.938873</td>\n",
-       "      <td>180.486734</td>\n",
+       "      <td>0.982671</td>\n",
+       "      <td>261.989623</td>\n",
+       "      <td>180.507339</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:04:46.446138</td>\n",
+       "      <td>0 days 00:04:30.437045</td>\n",
        "      <td>59</td>\n",
-       "      <td>0.205972</td>\n",
-       "      <td>72.277372</td>\n",
-       "      <td>564.734093</td>\n",
+       "      <td>0.218165</td>\n",
+       "      <td>56.699605</td>\n",
+       "      <td>443.038759</td>\n",
        "      <td>0.974582</td>\n",
        "      <td>317.519976</td>\n",
        "      <td>210.198793</td>\n",
@@ -2144,19 +2583,19 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:03.963151           461           116.319646   \n",
-       "featuretools    0 days 00:04:46.446138            59             0.205972   \n",
+       "getML: FastProp 0 days 00:00:04.769646           461            96.655712   \n",
+       "featuretools    0 days 00:04:30.437045            59             0.218165   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
-       "getML: FastProp            1.000000                        1.000000  0.982678   \n",
-       "featuretools              72.277372                      564.734093  0.974582   \n",
+       "getML: FastProp            1.000000                        1.000000  0.982671   \n",
+       "featuretools              56.699605                      443.038759  0.974582   \n",
        "\n",
        "                       rmse         mae  \n",
-       "getML: FastProp  261.938873  180.486734  \n",
+       "getML: FastProp  261.989623  180.507339  \n",
        "featuretools     317.519976  210.198793  "
       ]
      },
-     "execution_count": 28,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2167,7 +2606,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2195,7 +2634,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.18"
+   "version": "3.11.9"
   },
   "toc": {
    "base_numbering": 1,
diff --git a/fastprop_benchmark/occupancy_prop.ipynb b/fastprop_benchmark/occupancy_prop.ipynb
index cc09f4c..32c6e6b 100644
--- a/fastprop_benchmark/occupancy_prop.ipynb
+++ b/fastprop_benchmark/occupancy_prop.ipynb
@@ -24,6 +24,19 @@
     "- Population size: __32k__"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": [
+     "remove_cell_on_docs"
+    ]
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/occupancy_prop.ipynb\">\n",
+    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {
@@ -32,7 +45,7 @@
     }
    },
    "source": [
-    "# Background\n",
+    "## Background\n",
     "\n",
     "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
     "\n",
@@ -45,15 +58,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Analysis"
+    "## Analysis"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Table of contents\n",
-    "\n",
     "1. [Loading data](#1.-Loading-data)\n",
     "2. [Predictive modeling](#2.-Predictive-modeling)\n",
     "3. [Comparison](#3.-Comparison)"
@@ -75,51 +86,106 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML API version: 1.4.0\n",
-      "\n",
-      "getML engine is already running.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Loading hyperopts... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'occupancy'\n"
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML API version: 1.5.0\n",
+      "\n"
      ]
     }
    ],
    "source": [
-    "import datetime\n",
     "import os\n",
+    "import sys\n",
+    "\n",
     "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
     "from pathlib import Path\n",
     "\n",
-    "import sys\n",
-    "import time\n",
-    "from urllib import request\n",
-    "\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "import numpy as np\n",
     "import pandas as pd\n",
-    "\n",
-    "%matplotlib inline\n",
-    "\n",
     "import getml\n",
     "\n",
-    "print(f\"getML API version: {getml.__version__}\\n\")\n",
-    "\n",
-    "getml.engine.launch(home_directory=Path.home(), allow_remote_ips=True, token='token')\n",
+    "print(f\"getML API version: {getml.__version__}\\n\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML Engine is already running.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'occupancy'</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Connected to project \u001b[32m'occupancy'\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
     "getml.engine.set_project(\"occupancy\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
     "parent = Path(os.getcwd()).parent.as_posix()\n",
     "\n",
     "if parent not in sys.path:\n",
-    "   sys.path.append(parent) \n",
+    "    sys.path.append(parent)\n",
     "\n",
     "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
    ]
@@ -128,7 +194,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 1. Loading data\n"
+    "### 1. Loading data\n"
    ]
   },
   {
@@ -140,22 +206,109 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_train<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Loading population_train\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\n",
-      "Loading population_train...\n",
-      " 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Loading population_test...\n",
-      " 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Loading population_validation...\n",
-      " 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n"
+      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m554.6/554.6 kB\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_test<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Loading population_test\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m668.3/668.3 kB\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_validation<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Loading population_validation\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m186.5/186.5 kB\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     }
    ],
@@ -165,7 +318,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -186,7 +339,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {
     "tags": [
      "hide_input"
@@ -1338,7 +1491,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1351,7 +1504,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 2. Predictive modeling\n",
+    "### 2. Predictive modeling\n",
     "\n",
     "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
    ]
@@ -1360,7 +1513,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.1 Propositionalization with getML's FastProp"
+    "#### 2.1 Propositionalization with getML's FastProp"
    ]
   },
   {
@@ -1372,7 +1525,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -1508,7 +1661,7 @@
        "        \n",
        "      \n",
        "        \n",
-       "          <th class=\"int\">rows</th>\n",
+       "          <th class=\"str\">rows   </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -1532,7 +1685,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">8142</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1553,7 +1706,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">9753</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1574,7 +1727,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">2665</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1693,17 +1846,17 @@
        "container\n",
        "\n",
        "  population\n",
-       "      subset       name       rows   type\n",
-       "  0   test         data_all   8142   View\n",
-       "  1   train        data_all   9753   View\n",
-       "  2   validation   data_all   2665   View\n",
+       "      subset       name       rows      type\n",
+       "  0   test         data_all   unknown   View\n",
+       "  1   train        data_all   unknown   View\n",
+       "  2   validation   data_all   unknown   View\n",
        "\n",
        "  peripheral\n",
        "      name        rows   type     \n",
        "  0   data_all   20560   DataFrame"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1738,7 +1891,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1760,7 +1913,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1773,19 +1926,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1801,7 +1991,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1810,68 +2000,168 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 331 features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:1.177664\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"fastprop\"):\n",
-    "    pipe_fp_fl.fit(time_series.train)\n",
-    "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Trying 331 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m0%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
-    }
-   ],
-   "source": [
-    "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now we create a dedicated prediction pipeline and provide the fast prop features\n",
-    "(contrained in `fastprop_train` and `fastprop_test`.)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:01.035566.\n",
+      "\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "with benchmark(\"fastprop\"):\n",
+    "    pipe_fp_fl.fit(time_series.train)\n",
+    "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we create a dedicated prediction pipeline and provide the fast prop features\n",
+    "(contrained in `fastprop_train` and `fastprop_test`.)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
    "outputs": [],
    "source": [
     "predictor = getml.predictors.XGBoostClassifier()\n",
@@ -1883,27 +2173,144 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:05, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:5.153885\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:04.948174.\n",
       "\n"
      ]
     },
@@ -1934,7 +2341,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1947,18 +2354,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2028,7 +2448,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:26</td>\n",
+       "              <td class=\"datetime\">2024-09-05 17:46:11</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2057,7 +2477,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:26</td>\n",
+       "              <td class=\"datetime\">2024-09-05 17:46:12</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2077,7 +2497,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.044271</td>\n",
+       "              <td class=\"float\">0.044287</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -2087,11 +2507,11 @@
       ],
       "text/plain": [
        "    date time             set used         target      accuracy       auc   cross entropy\n",
-       "0   2024-02-21 16:13:26   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
-       "1   2024-02-21 16:13:26   fastprop_test    Occupancy     0.9888    0.9982        0.044271"
+       "0   2024-09-05 17:46:11   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
+       "1   2024-09-05 17:46:12   fastprop_test    Occupancy     0.9888    0.9982        0.044287"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2104,12 +2524,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.2 Propositionalization with featuretools"
+    "#### 2.2 Propositionalization with featuretools"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2119,7 +2539,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2132,7 +2552,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2156,7 +2576,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
     {
@@ -2165,7 +2585,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 262 features...\n",
-      "Time taken: 0h:8m:2.659652\n",
+      "Time taken: 0h:7m:18.06737\n",
       "\n"
      ]
     }
@@ -2194,7 +2614,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
@@ -2224,7 +2644,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 22,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2241,20 +2661,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2333,7 +2790,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 21,
+     "execution_count": 23,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2344,24 +2801,122 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n",
-      "To see the issues in full, run .check() on the pipeline.\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:04, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:4.049619\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:03.188635.\n",
       "\n"
      ]
     },
@@ -2392,7 +2947,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 24,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2403,18 +2958,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2484,7 +3052,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:29:04</td>\n",
+       "              <td class=\"datetime\">2024-09-05 17:59:45</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2496,7 +3064,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.9994</td>\n",
+       "              <td class=\"float\">0.9995</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2504,7 +3072,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.004997</td>\n",
+       "              <td class=\"float\">0.005065</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -2513,7 +3081,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:29:04</td>\n",
+       "              <td class=\"datetime\">2024-09-05 17:59:45</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2525,15 +3093,15 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.9886</td>\n",
+       "              <td class=\"float\">0.9885</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.9973</td>\n",
+       "              <td class=\"float\">0.9972</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.05008</td>\n",
+       "              <td class=\"float\">0.049236</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -2543,11 +3111,11 @@
       ],
       "text/plain": [
        "    date time             set used             target      accuracy       auc   cross entropy\n",
-       "0   2024-02-21 16:29:04   featuretools_train   Occupancy     0.9994    1.            0.004997\n",
-       "1   2024-02-21 16:29:04   featuretools_test    Occupancy     0.9886    0.9973        0.05008 "
+       "0   2024-09-05 17:59:45   featuretools_train   Occupancy     0.9995    1.            0.005065\n",
+       "1   2024-09-05 17:59:45   featuretools_test    Occupancy     0.9885    0.9972        0.049236"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2560,12 +3128,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.3 Propositionalization with tsfresh"
+    "#### 2.3 Propositionalization with tsfresh"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 26,
    "metadata": {
     "lines_to_next_cell": 2
    },
@@ -2574,9 +3142,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.42it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.95it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.80it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.45it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.26it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.59it/s]\n"
      ]
     },
     {
@@ -2584,7 +3152,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 65 features...\n",
-      "Time taken: 0h:0m:20.577953\n",
+      "Time taken: 0h:0m:16.408325\n",
       "\n"
      ]
     },
@@ -2592,9 +3160,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.87it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  6.77it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.10it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 19.16it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.80it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:03<00:00, 10.21it/s]\n"
      ]
     }
    ],
@@ -2622,7 +3190,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
@@ -2652,7 +3220,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2667,20 +3235,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2759,7 +3364,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 26,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2770,24 +3375,122 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 1 issues labeled WARNING.\n",
-      "To see the issues in full, run .check() on the pipeline.\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:02, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:1.896389\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:01.717708.\n",
       "\n"
      ]
     },
@@ -2818,7 +3521,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2829,18 +3532,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -2910,7 +3626,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:29:44</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:00:17</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2939,7 +3655,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:29:44</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:00:17</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2969,11 +3685,11 @@
       ],
       "text/plain": [
        "    date time             set used        target      accuracy       auc   cross entropy\n",
-       "0   2024-02-21 16:29:44   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
-       "1   2024-02-21 16:29:44   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
+       "0   2024-09-05 18:00:17   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
+       "1   2024-09-05 18:00:17   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2988,12 +3704,12 @@
     "lines_to_next_cell": 2
    },
    "source": [
-    "## 3. Comparison"
+    "### 3. Comparison"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 31,
    "metadata": {
     "lines_to_next_cell": 2
    },
@@ -3047,7 +3763,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 32,
    "metadata": {
     "tags": []
    },
@@ -3086,33 +3802,33 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:01.918944</td>\n",
+       "      <td>0 days 00:00:01.816878</td>\n",
        "      <td>289</td>\n",
-       "      <td>150.602410</td>\n",
+       "      <td>159.058374</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.988823</td>\n",
-       "      <td>0.998166</td>\n",
-       "      <td>0.044271</td>\n",
+       "      <td>0.998173</td>\n",
+       "      <td>0.044287</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:08:02.660716</td>\n",
+       "      <td>0 days 00:07:18.068960</td>\n",
        "      <td>103</td>\n",
-       "      <td>0.213400</td>\n",
-       "      <td>251.524128</td>\n",
-       "      <td>705.726807</td>\n",
-       "      <td>0.988578</td>\n",
-       "      <td>0.997259</td>\n",
-       "      <td>0.050080</td>\n",
+       "      <td>0.235123</td>\n",
+       "      <td>241.110829</td>\n",
+       "      <td>676.490695</td>\n",
+       "      <td>0.988455</td>\n",
+       "      <td>0.997207</td>\n",
+       "      <td>0.049236</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:20.578092</td>\n",
+       "      <td>0 days 00:00:16.408459</td>\n",
        "      <td>60</td>\n",
-       "      <td>2.915724</td>\n",
-       "      <td>10.723654</td>\n",
-       "      <td>51.651807</td>\n",
+       "      <td>3.656655</td>\n",
+       "      <td>9.031129</td>\n",
+       "      <td>43.498330</td>\n",
        "      <td>0.987718</td>\n",
        "      <td>0.997861</td>\n",
        "      <td>0.049359</td>\n",
@@ -3123,22 +3839,22 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:01.918944           289           150.602410   \n",
-       "featuretools    0 days 00:08:02.660716           103             0.213400   \n",
-       "tsfresh         0 days 00:00:20.578092            60             2.915724   \n",
+       "getML: FastProp 0 days 00:00:01.816878           289           159.058374   \n",
+       "featuretools    0 days 00:07:18.068960           103             0.235123   \n",
+       "tsfresh         0 days 00:00:16.408459            60             3.656655   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  accuracy  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.988823   \n",
-       "featuretools             251.524128                      705.726807  0.988578   \n",
-       "tsfresh                   10.723654                       51.651807  0.987718   \n",
+       "featuretools             241.110829                      676.490695  0.988455   \n",
+       "tsfresh                    9.031129                       43.498330  0.987718   \n",
        "\n",
        "                      auc  cross_entropy  \n",
-       "getML: FastProp  0.998166       0.044271  \n",
-       "featuretools     0.997259       0.050080  \n",
+       "getML: FastProp  0.998173       0.044287  \n",
+       "featuretools     0.997207       0.049236  \n",
        "tsfresh          0.997861       0.049359  "
       ]
      },
-     "execution_count": 30,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3149,7 +3865,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3179,7 +3895,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.18"
+   "version": "3.11.9"
   },
   "toc-autonumbering": false,
   "toc-showcode": false,
diff --git a/fastprop_benchmark/robot_prop.ipynb b/fastprop_benchmark/robot_prop.ipynb
index 727a317..8b25d87 100644
--- a/fastprop_benchmark/robot_prop.ipynb
+++ b/fastprop_benchmark/robot_prop.ipynb
@@ -17,11 +17,24 @@
     "- Population size: __15001__"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": [
+     "remove_cell_on_docs"
+    ]
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/robot_prop.ipynb\">\n",
+    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Background\n",
+    "## Background\n",
     "\n",
     "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
     "\n",
@@ -36,15 +49,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Analysis"
+    "## Analysis"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Table of contents\n",
-    "\n",
     "1. [Loading data](#1.-Loading-data)\n",
     "2. [Predictive modeling](#2.-Predictive-modeling)\n",
     "3. [Comparison](#3.-Comparison)"
@@ -66,51 +77,107 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML engine is already running.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'robot'\n"
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML API version: 1.5.0\n",
+      "\n"
      ]
     }
    ],
    "source": [
-    "import datetime\n",
     "import os\n",
+    "import sys\n",
+    "\n",
     "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
     "from pathlib import Path\n",
     "\n",
-    "import sys\n",
-    "import time\n",
-    "from urllib import request\n",
-    "\n",
-    "import getml\n",
-    "import getml.data as data\n",
-    "import getml.data.roles as roles\n",
-    "import getml.database as database\n",
-    "import getml.engine as engine\n",
-    "import getml.feature_learning.aggregations as agg\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
     "import numpy as np\n",
     "import pandas as pd\n",
-    "from IPython.display import Image\n",
-    "\n",
-    "%matplotlib inline\n",
+    "import getml\n",
     "\n",
-    "getml.engine.launch(home_directory=Path.home(), allow_remote_ips=True, token='token')\n",
+    "print(f\"getML API version: {getml.__version__}\\n\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "getML Engine is already running.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'robot'</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Connected to project \u001b[32m'robot'\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
     "getml.engine.set_project(\"robot\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
     "parent = Path(os.getcwd()).parent.as_posix()\n",
     "\n",
     "if parent not in sys.path:\n",
-    "   sys.path.append(parent) \n",
+    "    sys.path.append(parent)\n",
     "\n",
     "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
    ]
@@ -119,27 +186,42 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 1. Loading data"
+    "### 1. Loading data"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.1 Download from source\n"
+    "#### 1.1 Download from source\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Downloading <span style=\"color: #0000ff; text-decoration-color: #0000ff; text-decoration: underline\">https://static.getml.com/datasets/robotarm/robot-demo.csv</span> to \n",
+       "<span style=\"color: #800080; text-decoration-color: #800080\">/tmp/getml/static.getml.com/datasets/robotarm/</span><span style=\"color: #ff00ff; text-decoration-color: #ff00ff\">robot-demo.csv...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Downloading \u001b[4;94mhttps://static.getml.com/datasets/robotarm/robot-demo.csv\u001b[0m to \n",
+       "\u001b[35m/tmp/getml/static.getml.com/datasets/robotarm/\u001b[0m\u001b[95mrobot-demo.csv...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Downloading robot-demo.csv...\n",
-      " 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n"
+      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m14.7/14.7 MB\u001b[0m • \u001b[33m00:00\u001b[0m4.7 MB\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
     }
    ],
@@ -151,7 +233,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -14814,7 +14896,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 4,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -14827,7 +14909,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.2 Prepare data\n",
+    "#### 1.2 Prepare data\n",
     "\n",
     "The force vector consists of three component (*f_x*, *f_y* and *f_z*), meaning that we have three targets. For this comparison, we only predict the first component (*f_x*). \n",
     "\n",
@@ -14836,7 +14918,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -14855,7 +14937,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -29518,7 +29600,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -29531,14 +29613,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1.3 Separate data into a training and testing set\n",
+    "#### 1.3 Separate data into a training and testing set\n",
     "\n",
     "We also want to separate the data set into a training and testing set. We do so by using the first 10,500 measurements for training and then using the remainder for testing."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -29585,7 +29667,7 @@
        "    <tr>\n",
        "      \n",
        "        \n",
-       "          <th>     </th>\n",
+       "          <th>  </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -29655,27 +29737,27 @@
        "</table>\n",
        "\n",
        "  <p>\n",
-       "    15001 rows<br />\n",
+       "    unknown number of  rows<br />\n",
        "    \n",
        "    type: StringColumnView<br />\n",
        "    \n",
        "  </p>\n"
       ],
       "text/plain": [
-       "             \n",
-       "    0   train\n",
-       "    1   train\n",
-       "    2   train\n",
-       "    3   train\n",
-       "    4   train\n",
-       "        ...  \n",
+       "          \n",
+       " 0   train\n",
+       " 1   train\n",
+       " 2   train\n",
+       " 3   train\n",
+       " 4   train\n",
+       "     ...  \n",
        "\n",
        "\n",
-       "15001 rows\n",
+       "unknown number of  rows\n",
        "type: StringColumnView"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -29687,7 +29769,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -29823,7 +29905,7 @@
        "        \n",
        "      \n",
        "        \n",
-       "          <th class=\"int\"> rows</th>\n",
+       "          <th class=\"str\">rows   </th>\n",
        "        \n",
        "      \n",
        "        \n",
@@ -29847,7 +29929,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">4501</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -29868,7 +29950,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"int\">10500</td>\n",
+       "              <td class=\"str\">unknown</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -29986,16 +30068,16 @@
        "container\n",
        "\n",
        "  population\n",
-       "      subset   name        rows   type\n",
-       "  0   test     data_all    4501   View\n",
-       "  1   train    data_all   10500   View\n",
+       "      subset   name       rows      type\n",
+       "  0   test     data_all   unknown   View\n",
+       "  1   train    data_all   unknown   View\n",
        "\n",
        "  peripheral\n",
        "      name        rows   type\n",
        "  0   data_all   15001   View"
       ]
      },
-     "execution_count": 8,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30016,14 +30098,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 2. Predictive modeling\n",
+    "### 2. Predictive modeling\n",
     "\n",
-    "### 2.1 Propositionalization with getML's FastProp"
+    "#### 2.1 Propositionalization with getML's FastProp"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30034,7 +30116,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30047,19 +30129,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -30068,7 +30187,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30077,28 +30196,115 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 134 features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:0.012022\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Trying 134 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:00.037119.\n",
       "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -30109,18 +30315,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -30129,7 +30348,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30142,20 +30361,57 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 5 issues labeled WARNING.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">5</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m5\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -30221,7 +30477,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_123&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_126&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30238,7 +30494,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_125&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_127&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30255,7 +30511,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_128&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_129&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30272,7 +30528,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_129&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_130&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30306,7 +30562,7 @@
        "4   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 16,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30317,24 +30573,122 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "The pipeline check generated 0 issues labeled INFO and 5 issues labeled WARNING.\n",
-      "To see the issues in full, run .check() on the pipeline.\n",
-      "\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:05, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:4.613744\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">5</span> issues labeled WARNING.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m5\u001b[0m issues labeled WARNING.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:04.133426.\n",
       "\n"
      ]
     },
@@ -30365,7 +30719,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 19,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30376,18 +30730,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -30457,7 +30824,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:41</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:17:16</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30486,7 +30853,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:13:41</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:17:17</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30516,11 +30883,11 @@
       ],
       "text/plain": [
        "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:13:41   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
-       "1   2024-02-21 16:13:41   fastprop_test    f_x       0.5515    0.7236     0.9951"
+       "0   2024-09-05 18:17:16   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
+       "1   2024-09-05 18:17:17   fastprop_test    f_x       0.5515    0.7236     0.9951"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 20,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30533,12 +30900,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.2 Propositionalization with featuretools"
+    "#### 2.2 Propositionalization with featuretools"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30548,7 +30915,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30569,7 +30936,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
@@ -30820,7 +31187,7 @@
        "[10500 rows x 13 columns]"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 23,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30831,7 +31198,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30847,7 +31214,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -30856,7 +31223,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 442 features...\n",
-      "Time taken: 0h:16m:25.820024\n",
+      "Time taken: 0h:14m:35.631365\n",
       "\n"
      ]
     }
@@ -30870,7 +31237,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -30958,10 +31325,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
        "      <td>-11.0300</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:00</td>\n",
@@ -30982,10 +31349,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
        "      <td>-10.8480</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:01</td>\n",
@@ -31006,10 +31373,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
        "      <td>-10.6660</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:02</td>\n",
@@ -31030,10 +31397,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
        "      <td>-10.5070</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:03</td>\n",
@@ -31054,10 +31421,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
-       "      <td>1.708532e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
+       "      <td>1.725560e+09</td>\n",
        "      <td>-10.4130</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:04</td>\n",
@@ -31102,10 +31469,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
        "      <td>-9.7673</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:55</td>\n",
@@ -31126,10 +31493,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
        "      <td>-9.9200</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:56</td>\n",
@@ -31150,10 +31517,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
        "      <td>-9.7743</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:57</td>\n",
@@ -31174,10 +31541,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
        "      <td>-8.6109</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:58</td>\n",
@@ -31198,10 +31565,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
-       "      <td>1.708522e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
+       "      <td>1.725550e+09</td>\n",
        "      <td>-8.4345</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:59</td>\n",
@@ -31326,59 +31693,59 @@
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 4)  \\\n",
        "_featuretools_index                                          \n",
-       "0                                             1.708532e+09   \n",
-       "1                                             1.708532e+09   \n",
-       "2                                             1.708532e+09   \n",
-       "3                                             1.708532e+09   \n",
-       "4                                             1.708532e+09   \n",
+       "0                                             1.725560e+09   \n",
+       "1                                             1.725560e+09   \n",
+       "2                                             1.725560e+09   \n",
+       "3                                             1.725560e+09   \n",
+       "4                                             1.725560e+09   \n",
        "...                                                    ...   \n",
-       "10495                                         1.708522e+09   \n",
-       "10496                                         1.708522e+09   \n",
-       "10497                                         1.708522e+09   \n",
-       "10498                                         1.708522e+09   \n",
-       "10499                                         1.708522e+09   \n",
+       "10495                                         1.725550e+09   \n",
+       "10496                                         1.725550e+09   \n",
+       "10497                                         1.725550e+09   \n",
+       "10498                                         1.725550e+09   \n",
+       "10499                                         1.725550e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 30)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.708532e+09   \n",
-       "1                                              1.708532e+09   \n",
-       "2                                              1.708532e+09   \n",
-       "3                                              1.708532e+09   \n",
-       "4                                              1.708532e+09   \n",
+       "0                                              1.725560e+09   \n",
+       "1                                              1.725560e+09   \n",
+       "2                                              1.725560e+09   \n",
+       "3                                              1.725560e+09   \n",
+       "4                                              1.725560e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.708522e+09   \n",
-       "10496                                          1.708522e+09   \n",
-       "10497                                          1.708522e+09   \n",
-       "10498                                          1.708522e+09   \n",
-       "10499                                          1.708522e+09   \n",
+       "10495                                          1.725550e+09   \n",
+       "10496                                          1.725550e+09   \n",
+       "10497                                          1.725550e+09   \n",
+       "10498                                          1.725550e+09   \n",
+       "10499                                          1.725550e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 77)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.708532e+09   \n",
-       "1                                              1.708532e+09   \n",
-       "2                                              1.708532e+09   \n",
-       "3                                              1.708532e+09   \n",
-       "4                                              1.708532e+09   \n",
+       "0                                              1.725560e+09   \n",
+       "1                                              1.725560e+09   \n",
+       "2                                              1.725560e+09   \n",
+       "3                                              1.725560e+09   \n",
+       "4                                              1.725560e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.708522e+09   \n",
-       "10496                                          1.708522e+09   \n",
-       "10497                                          1.708522e+09   \n",
-       "10498                                          1.708522e+09   \n",
-       "10499                                          1.708522e+09   \n",
+       "10495                                          1.725550e+09   \n",
+       "10496                                          1.725550e+09   \n",
+       "10497                                          1.725550e+09   \n",
+       "10498                                          1.725550e+09   \n",
+       "10499                                          1.725550e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 7)      f_x  id  \\\n",
        "_featuretools_index                                                       \n",
-       "0                                             1.708532e+09 -11.0300   1   \n",
-       "1                                             1.708532e+09 -10.8480   1   \n",
-       "2                                             1.708532e+09 -10.6660   1   \n",
-       "3                                             1.708532e+09 -10.5070   1   \n",
-       "4                                             1.708532e+09 -10.4130   1   \n",
+       "0                                             1.725560e+09 -11.0300   1   \n",
+       "1                                             1.725560e+09 -10.8480   1   \n",
+       "2                                             1.725560e+09 -10.6660   1   \n",
+       "3                                             1.725560e+09 -10.5070   1   \n",
+       "4                                             1.725560e+09 -10.4130   1   \n",
        "...                                                    ...      ...  ..   \n",
-       "10495                                         1.708522e+09  -9.7673   1   \n",
-       "10496                                         1.708522e+09  -9.9200   1   \n",
-       "10497                                         1.708522e+09  -9.7743   1   \n",
-       "10498                                         1.708522e+09  -8.6109   1   \n",
-       "10499                                         1.708522e+09  -8.4345   1   \n",
+       "10495                                         1.725550e+09  -9.7673   1   \n",
+       "10496                                         1.725550e+09  -9.9200   1   \n",
+       "10497                                         1.725550e+09  -9.7743   1   \n",
+       "10498                                         1.725550e+09  -8.6109   1   \n",
+       "10499                                         1.725550e+09  -8.4345   1   \n",
        "\n",
        "                                     ds  \n",
        "_featuretools_index                      \n",
@@ -31397,7 +31764,7 @@
        "[10500 rows x 203 columns]"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 26,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31408,7 +31775,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31429,7 +31796,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31444,7 +31811,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
@@ -31474,7 +31841,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31491,23 +31858,97 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:05, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:5.114944\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:04.440980.\n",
       "\n"
      ]
     },
@@ -31538,7 +31979,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31549,18 +31990,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -31630,7 +32084,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:37:13</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:38:13</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31642,11 +32096,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.4387</td>\n",
+       "              <td class=\"float\">0.4395</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.5826</td>\n",
+       "              <td class=\"float\">0.5835</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31659,7 +32113,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:37:14</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:38:13</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31671,11 +32125,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.5718</td>\n",
+       "              <td class=\"float\">0.572</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.7481</td>\n",
+       "              <td class=\"float\">0.7486</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31689,11 +32143,11 @@
       ],
       "text/plain": [
        "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:37:13   featuretools_train   f_x       0.4387    0.5826     0.9962\n",
-       "1   2024-02-21 16:37:14   featuretools_test    f_x       0.5718    0.7481     0.9948"
+       "0   2024-09-05 18:38:13   featuretools_train   f_x       0.4395    0.5835     0.9962\n",
+       "1   2024-09-05 18:38:13   featuretools_test    f_x       0.572     0.7486     0.9948"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31706,12 +32160,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2.3 Propositionalization with tsfresh"
+    "#### 2.3 Propositionalization with tsfresh"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31726,16 +32180,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.71it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:13<00:00,  3.01it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:13<00:00,  3.08it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.96it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:10<00:00,  3.95it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:10<00:00,  3.94it/s]\n"
      ]
     },
     {
@@ -31743,7 +32197,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 130 features...\n",
-      "Time taken: 0h:0m:34.618933\n",
+      "Time taken: 0h:0m:26.245796\n",
       "\n"
      ]
     },
@@ -31751,9 +32205,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 26.85it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.43it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.22it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 35.68it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.43it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.14it/s]\n"
      ]
     }
    ],
@@ -31766,7 +32220,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31787,7 +32241,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31798,7 +32252,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [
     {
@@ -31828,7 +32282,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 34,
+     "execution_count": 36,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31843,19 +32297,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Checking... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -31864,22 +32355,96 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Checking data model\u001b[33m...\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Checking data model...\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "OK.\n",
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:04, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:4.40497\n",
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "OK.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[?25h"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "Trained pipeline.\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time taken: 0:00:03.945516.\n",
       "\n"
      ]
     },
@@ -31910,7 +32475,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 38,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31921,18 +32486,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
+      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
+      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[?25h"
      ]
     },
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
     {
      "data": {
       "text/html": [
@@ -32002,7 +32580,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:38:08</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:38:57</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32031,7 +32609,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-02-21 16:38:08</td>\n",
+       "              <td class=\"datetime\">2024-09-05 18:38:57</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32061,11 +32639,11 @@
       ],
       "text/plain": [
        "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-02-21 16:38:08   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
-       "1   2024-02-21 16:38:08   tsfresh_test    f_x       0.5986    0.7906     0.9938"
+       "0   2024-09-05 18:38:57   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
+       "1   2024-09-05 18:38:57   tsfresh_test    f_x       0.5986    0.7906     0.9938"
       ]
      },
-     "execution_count": 37,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32078,12 +32656,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 3. Comparison"
+    "### 3. Comparison"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32131,7 +32709,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [
     {
@@ -32168,9 +32746,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:00.278417</td>\n",
+       "      <td>0 days 00:00:00.424167</td>\n",
        "      <td>134</td>\n",
-       "      <td>481.231954</td>\n",
+       "      <td>315.955766</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.995059</td>\n",
@@ -32179,22 +32757,22 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:16:25.821720</td>\n",
+       "      <td>0 days 00:14:35.632221</td>\n",
        "      <td>158</td>\n",
-       "      <td>0.160272</td>\n",
-       "      <td>3540.810080</td>\n",
-       "      <td>3002.588065</td>\n",
-       "      <td>0.994791</td>\n",
-       "      <td>0.748083</td>\n",
-       "      <td>0.571803</td>\n",
+       "      <td>0.180441</td>\n",
+       "      <td>2064.357248</td>\n",
+       "      <td>1751.019273</td>\n",
+       "      <td>0.994803</td>\n",
+       "      <td>0.748614</td>\n",
+       "      <td>0.572019</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:34.619104</td>\n",
+       "      <td>0 days 00:00:26.245968</td>\n",
        "      <td>120</td>\n",
-       "      <td>3.466289</td>\n",
-       "      <td>124.342637</td>\n",
-       "      <td>138.832050</td>\n",
+       "      <td>4.572139</td>\n",
+       "      <td>61.876497</td>\n",
+       "      <td>69.104581</td>\n",
        "      <td>0.993836</td>\n",
        "      <td>0.790602</td>\n",
        "      <td>0.598600</td>\n",
@@ -32205,22 +32783,22 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:00.278417           134           481.231954   \n",
-       "featuretools    0 days 00:16:25.821720           158             0.160272   \n",
-       "tsfresh         0 days 00:00:34.619104           120             3.466289   \n",
+       "getML: FastProp 0 days 00:00:00.424167           134           315.955766   \n",
+       "featuretools    0 days 00:14:35.632221           158             0.180441   \n",
+       "tsfresh         0 days 00:00:26.245968           120             4.572139   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.995059   \n",
-       "featuretools            3540.810080                     3002.588065  0.994791   \n",
-       "tsfresh                  124.342637                      138.832050  0.993836   \n",
+       "featuretools            2064.357248                     1751.019273  0.994803   \n",
+       "tsfresh                   61.876497                       69.104581  0.993836   \n",
        "\n",
        "                     rmse       mae  \n",
        "getML: FastProp  0.723622  0.551521  \n",
-       "featuretools     0.748083  0.571803  \n",
+       "featuretools     0.748614  0.572019  \n",
        "tsfresh          0.790602  0.598600  "
       ]
      },
-     "execution_count": 39,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32231,7 +32809,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32255,7 +32833,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.18"
+   "version": "3.11.9"
   }
  },
  "nbformat": 4,
diff --git a/utils/ft_time_series_builder.py b/utils/ft_time_series_builder.py
index fa67ab1..031637e 100644
--- a/utils/ft_time_series_builder.py
+++ b/utils/ft_time_series_builder.py
@@ -96,6 +96,7 @@ def _roll_data_frame(data_frame, column_id, time_stamp, horizon, memory):
 def _hide_warnings(func):
     def wrapper(*args, **kwargs):
         with warnings.catch_warnings():
+            warnings.simplefilter(action="ignore", category=FutureWarning)
             warnings.filterwarnings(
                 'ignore',
                 category=RuntimeWarning,

From 8f586f3494b43765f47b7105b80b1b249a02486a Mon Sep 17 00:00:00 2001
From: Alexandros Ladas <alexandros@getml.com>
Date: Thu, 12 Sep 2024 17:59:52 +0200
Subject: [PATCH 2/3] Updated fastprop benchmark notebooks to getml 1.5.0

---
 fastprop_benchmark/air_pollution_prop.ipynb   | 483 ++++--------
 .../comparisons/air_pollution.csv             |   6 +-
 fastprop_benchmark/comparisons/dodgers.csv    |   6 +-
 .../comparisons/interstate94.csv              |   4 +-
 fastprop_benchmark/comparisons/occupancy.csv  |   6 +-
 fastprop_benchmark/comparisons/robot.csv      |   6 +-
 fastprop_benchmark/dodgers_prop.ipynb         | 499 ++++---------
 fastprop_benchmark/interstate94_prop.ipynb    | 393 +++-------
 fastprop_benchmark/occupancy_prop.ipynb       | 611 ++++------------
 fastprop_benchmark/robot_prop.ipynb           | 692 ++++++------------
 10 files changed, 731 insertions(+), 1975 deletions(-)

diff --git a/fastprop_benchmark/air_pollution_prop.ipynb b/fastprop_benchmark/air_pollution_prop.ipynb
index b13be43..59b0f7e 100644
--- a/fastprop_benchmark/air_pollution_prop.ipynb
+++ b/fastprop_benchmark/air_pollution_prop.ipynb
@@ -114,7 +114,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -125,7 +125,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -155,7 +155,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -171,7 +171,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -186,7 +186,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -210,7 +210,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -230,7 +230,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -271,7 +271,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -294,7 +294,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -305,7 +305,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -326,7 +326,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -516,7 +516,7 @@
        "[41757 rows x 9 columns]"
       ]
      },
-     "execution_count": 12,
+     "execution_count": 13,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -534,7 +534,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [
     {
@@ -542,32 +542,19 @@
      "output_type": "stream",
      "text": [
       "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-community-edition-linux...\n",
-      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240905115118.log.\n",
-      "\u001b[2K  Loading pipelines... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 83%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240912153535.log.\n",
+      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'air_pollution'</span>\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'air_pollution'</span>.\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "Connected to project \u001b[32m'air_pollution'\u001b[0m\n"
+       "Connected to project \u001b[32m'air_pollution'\u001b[0m.\n"
       ]
      },
      "metadata": {},
@@ -581,7 +568,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -598,7 +585,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
@@ -2071,7 +2058,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2087,7 +2074,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
@@ -2224,7 +2211,7 @@
        "type: StringColumnView"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2252,7 +2239,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -2560,7 +2547,7 @@
        "  0   full   41757   DataFrame"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2579,7 +2566,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
@@ -2609,7 +2596,7 @@
        "         tags=['memory: 1d', 'simple features'])"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 19,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2632,7 +2619,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
     {
@@ -2652,24 +2639,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2690,7 +2664,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2699,7 +2673,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
@@ -2719,23 +2693,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2753,24 +2714,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Trying 331 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:07\u001b[0m0%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2788,26 +2736,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:07.707196.\n",
+      "Time taken: 0:00:04.439985.\n",
       "\n",
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:07\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2818,31 +2753,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2851,7 +2773,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2864,7 +2786,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -2884,24 +2806,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2919,24 +2828,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:27\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:15\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2954,7 +2850,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:27.386765.\n",
+      "Time taken: 0:00:15.893621.\n",
       "\n"
      ]
     },
@@ -2985,7 +2881,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2996,31 +2892,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:27\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3090,7 +2973,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 11:52:05</td>\n",
+       "              <td class=\"datetime\">2024-09-12 15:36:02</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3119,7 +3002,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 11:52:06</td>\n",
+       "              <td class=\"datetime\">2024-09-12 15:36:02</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3149,11 +3032,11 @@
       ],
       "text/plain": [
        "    date time             set used         target        mae       rmse   rsquared\n",
-       "0   2024-09-05 11:52:05   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
-       "1   2024-09-05 11:52:06   fastprop_test    pm2.5     44.2526    63.4191     0.5462"
+       "0   2024-09-12 15:36:02   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
+       "1   2024-09-12 15:36:02   fastprop_test    pm2.5     44.2526    63.4191     0.5462"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 26,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3173,7 +3056,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3189,7 +3072,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
@@ -3198,7 +3081,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 298 features...\n",
-      "Time taken: 0h:33m:59.375927\n",
+      "Time taken: 0h:29m:33.992379\n",
       "\n"
      ]
     }
@@ -3212,7 +3095,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3226,7 +3109,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3245,7 +3128,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
@@ -3275,7 +3158,7 @@
        "         tags=['featuretools', 'memory: 1d'])"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3292,7 +3175,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
@@ -3312,24 +3195,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3350,7 +3220,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
@@ -3370,23 +3240,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3404,24 +3261,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:09\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:10\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3439,7 +3283,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:09.898496.\n",
+      "Time taken: 0:00:10.147840.\n",
       "\n"
      ]
     },
@@ -3470,7 +3314,7 @@
        "         tags=['featuretools', 'memory: 1d'])"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3481,31 +3325,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:09\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3575,7 +3406,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 12:34:12</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:13:39</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3604,7 +3435,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 12:34:12</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:13:39</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3634,11 +3465,11 @@
       ],
       "text/plain": [
        "    date time             set used                target        mae       rmse   rsquared\n",
-       "0   2024-09-05 12:34:12   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
-       "1   2024-09-05 12:34:12   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
+       "0   2024-09-12 16:13:39   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
+       "1   2024-09-12 16:13:39   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
       ]
      },
-     "execution_count": 33,
+     "execution_count": 34,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3669,7 +3500,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [
     {
@@ -3859,7 +3690,7 @@
        "[33096 rows x 9 columns]"
       ]
      },
-     "execution_count": 34,
+     "execution_count": 35,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3870,16 +3701,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:20<00:00,  1.92it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.71it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.74it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:22<00:00,  1.77it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:22<00:00,  1.79it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:22<00:00,  1.75it/s]\n"
      ]
     },
     {
@@ -3887,7 +3718,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 78 features...\n",
-      "Time taken: 0h:1m:13.903742\n",
+      "Time taken: 0h:1m:15.803921\n",
       "\n"
      ]
     },
@@ -3895,9 +3726,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 16.30it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.04it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.02it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 16.82it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.35it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.05it/s]\n"
      ]
     }
    ],
@@ -3923,7 +3754,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3942,7 +3773,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3968,7 +3799,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
@@ -3998,7 +3829,7 @@
        "         tags=['tsfresh', 'memory: 1d'])"
       ]
      },
-     "execution_count": 38,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4015,7 +3846,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
     {
@@ -4035,24 +3866,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -4073,7 +3891,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [
     {
@@ -4093,23 +3911,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -4127,24 +3932,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:05\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:05\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -4162,7 +3954,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:05.505439.\n",
+      "Time taken: 0:00:05.349890.\n",
       "\n"
      ]
     },
@@ -4193,7 +3985,7 @@
        "         tags=['tsfresh', 'memory: 1d'])"
       ]
      },
-     "execution_count": 40,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4204,31 +3996,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:05\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -4298,7 +4077,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 12:35:50</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:15:17</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -4327,7 +4106,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 12:35:50</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:15:18</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -4357,11 +4136,11 @@
       ],
       "text/plain": [
        "    date time             set used           target        mae       rmse   rsquared\n",
-       "0   2024-09-05 12:35:50   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
-       "1   2024-09-05 12:35:50   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
+       "0   2024-09-12 16:15:17   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
+       "1   2024-09-12 16:15:18   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
       ]
      },
-     "execution_count": 41,
+     "execution_count": 42,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4379,7 +4158,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 43,
    "metadata": {},
    "outputs": [
     {
@@ -4416,9 +4195,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:12.524776</td>\n",
+       "      <td>0 days 00:00:07.285902</td>\n",
        "      <td>289</td>\n",
-       "      <td>23.074438</td>\n",
+       "      <td>39.665225</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>44.252635</td>\n",
@@ -4427,22 +4206,22 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:33:59.377855</td>\n",
+       "      <td>0 days 00:29:33.994416</td>\n",
        "      <td>114</td>\n",
-       "      <td>0.055899</td>\n",
-       "      <td>162.827491</td>\n",
-       "      <td>412.785062</td>\n",
+       "      <td>0.064262</td>\n",
+       "      <td>243.483156</td>\n",
+       "      <td>617.244655</td>\n",
        "      <td>43.915071</td>\n",
        "      <td>62.567175</td>\n",
        "      <td>0.559369</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:01:13.904058</td>\n",
+       "      <td>0 days 00:01:15.804228</td>\n",
        "      <td>72</td>\n",
-       "      <td>0.974236</td>\n",
-       "      <td>5.900629</td>\n",
-       "      <td>23.684642</td>\n",
+       "      <td>0.949816</td>\n",
+       "      <td>10.404234</td>\n",
+       "      <td>41.760977</td>\n",
        "      <td>47.110594</td>\n",
        "      <td>66.599982</td>\n",
        "      <td>0.501524</td>\n",
@@ -4453,14 +4232,14 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:12.524776           289            23.074438   \n",
-       "featuretools    0 days 00:33:59.377855           114             0.055899   \n",
-       "tsfresh         0 days 00:01:13.904058            72             0.974236   \n",
+       "getML: FastProp 0 days 00:00:07.285902           289            39.665225   \n",
+       "featuretools    0 days 00:29:33.994416           114             0.064262   \n",
+       "tsfresh         0 days 00:01:15.804228            72             0.949816   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  \\\n",
        "getML: FastProp            1.000000                        1.000000   \n",
-       "featuretools             162.827491                      412.785062   \n",
-       "tsfresh                    5.900629                       23.684642   \n",
+       "featuretools             243.483156                      617.244655   \n",
+       "tsfresh                   10.404234                       41.760977   \n",
        "\n",
        "                       mae       rmse  rsquared  \n",
        "getML: FastProp  44.252635  63.419113  0.546164  \n",
@@ -4468,7 +4247,7 @@
        "tsfresh          47.110594  66.599982  0.501524  "
       ]
      },
-     "execution_count": 42,
+     "execution_count": 43,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4520,7 +4299,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 44,
    "metadata": {
     "tags": []
    },
diff --git a/fastprop_benchmark/comparisons/air_pollution.csv b/fastprop_benchmark/comparisons/air_pollution.csv
index 19ab44f..62e1ecb 100644
--- a/fastprop_benchmark/comparisons/air_pollution.csv
+++ b/fastprop_benchmark/comparisons/air_pollution.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,mae,rmse,rsquared
-getML: FastProp,0 days 00:00:08.637513,289,33.45824411134903,1.0,1.0,44.2526350869082,63.41911260454679,0.5461637964380341
-featuretools,0 days 00:34:01.839960,114,0.055831995273039954,236.3921142578888,599.2664949143469,43.91507134716814,62.56717509580391,0.5593690691075496
-tsfresh,0 days 00:01:30.572946,72,0.7949390997155709,10.485998226572857,42.089065845824415,47.11059400765295,66.59998183642728,0.5015240507138585
+getML: FastProp,0 days 00:00:07.285902,289,39.66522549680695,1.0,1.0,44.2526350869082,63.41911260454679,0.5461637964380341
+featuretools,0 days 00:29:33.994416,114,0.06426175612599289,243.4831563751475,617.2446551108643,43.91507134716814,62.56717509580391,0.5593690691075496
+tsfresh,0 days 00:01:15.804228,72,0.9498155458210015,10.40423382032863,41.76097735115624,47.11059400765295,66.59998183642728,0.5015240507138585
diff --git a/fastprop_benchmark/comparisons/dodgers.csv b/fastprop_benchmark/comparisons/dodgers.csv
index 30cac3c..6b7ad12 100644
--- a/fastprop_benchmark/comparisons/dodgers.csv
+++ b/fastprop_benchmark/comparisons/dodgers.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:09.406415,526,55.9190292456523,1.0,1.0,0.674739861410577,7.824273388224941,5.615137721839041
-featuretools,0 days 00:09:19.754041,59,0.10540343322170759,59.50769139996481,530.523793546944,0.6497682308113026,8.500887091670293,6.08627656339764
-tsfresh,0 days 00:00:46.115063,12,0.2602186565327113,4.902512062246881,214.8924677067606,0.5778110797835911,8.913407825293008,6.788610043264059
+getML: FastProp,0 days 00:00:12.036174,526,43.7024735600035,1.0,1.0,0.674739861410577,7.824273388224941,5.615137721839041
+featuretools,0 days 00:08:56.482944,59,0.10997554034007297,44.572548053891545,397.38357661043614,0.6497682308113026,8.500887091670293,6.08627656339764
+tsfresh,0 days 00:00:32.318518,12,0.3713041315010712,2.685115552500321,117.69993881653701,0.5778110797835911,8.913407825293008,6.788610043264059
diff --git a/fastprop_benchmark/comparisons/interstate94.csv b/fastprop_benchmark/comparisons/interstate94.csv
index 2749d00..118c1ba 100644
--- a/fastprop_benchmark/comparisons/interstate94.csv
+++ b/fastprop_benchmark/comparisons/interstate94.csv
@@ -1,3 +1,3 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:03.963151,461,116.31964638827498,1.0,1.0,0.9826778318692238,261.9388731680907,180.4867341518402
-featuretools,0 days 00:04:46.446138,59,0.205972417409695,72.2773717176055,564.7340932883563,0.9745821357660296,317.51997565190663,210.1987933667501
+getML: FastProp,0 days 00:00:04.800454,461,96.03380389897244,1.0,1.0,0.9826778318692238,261.9388731680907,180.4867341518402
+featuretools,0 days 00:04:33.370925,59,0.21582395326461787,56.946889815005,444.9636031883223,0.9745821357660296,317.51997565190663,210.1987933667501
diff --git a/fastprop_benchmark/comparisons/occupancy.csv b/fastprop_benchmark/comparisons/occupancy.csv
index 5e528dc..fc554fb 100644
--- a/fastprop_benchmark/comparisons/occupancy.csv
+++ b/fastprop_benchmark/comparisons/occupancy.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,accuracy,auc,cross_entropy
-getML: FastProp,0 days 00:00:01.918944,289,150.6024096385542,1.0,1.0,0.9888233849177106,0.9981658593326046,0.044271389667191616
-featuretools,0 days 00:08:02.660716,103,0.21340043781233822,251.52412785365283,705.7268072289156,0.9885777450257922,0.9972594581153477,0.0500798548174994
-tsfresh,0 days 00:00:20.578092,60,2.915723915933848,10.723654259842913,51.651807228915665,0.9877180054040776,0.9978609436533588,0.049358601666829864
+getML: FastProp,0 days 00:00:01.854957,289,155.78750584203146,1.0,1.0,0.9888233849177106,0.9981659495576909,0.0442127968034932
+featuretools,0 days 00:07:28.129456,103,0.2298443195470412,241.58482164276583,677.7957625798412,0.988454925079833,0.9972073080154513,0.04923589235646241
+tsfresh,0 days 00:00:17.070465,60,3.51483965301503,9.202620330282588,44.32279171210469,0.9877180054040776,0.9978609436533588,0.04935860166671465
diff --git a/fastprop_benchmark/comparisons/robot.csv b/fastprop_benchmark/comparisons/robot.csv
index 947d254..00aec93 100644
--- a/fastprop_benchmark/comparisons/robot.csv
+++ b/fastprop_benchmark/comparisons/robot.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:00.278417,134,481.23195380173246,1.0,1.0,0.9950588372738192,0.7236222637718679,0.5515208330459932
-featuretools,0 days 00:16:25.821720,158,0.1602723861256683,3540.8100798442624,3002.588065447546,0.9947913268168053,0.7480833930239688,0.5718029095391902
-tsfresh,0 days 00:00:34.619104,120,3.466288610122256,124.34263712345151,138.8320500481232,0.9938361598092297,0.7906022324089085,0.5986000367767544
+getML: FastProp,0 days 00:00:00.434939,134,308.07147258163894,1.0,1.0,0.9950584802720546,0.7236738045405348,0.5515934724757293
+featuretools,0 days 00:14:36.193549,158,0.1803254477616112,2014.5205396618837,1708.4192852741837,0.9948315985902971,0.7484233883213018,0.571029394526552
+tsfresh,0 days 00:00:31.641862,120,3.7924469626292274,72.75011438385613,81.23290203327171,0.9938361598092297,0.7906022324089085,0.5986000367767544
diff --git a/fastprop_benchmark/dodgers_prop.ipynb b/fastprop_benchmark/dodgers_prop.ipynb
index b2cefb7..f65e8bd 100644
--- a/fastprop_benchmark/dodgers_prop.ipynb
+++ b/fastprop_benchmark/dodgers_prop.ipynb
@@ -112,7 +112,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -123,7 +123,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -137,39 +137,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML Engine is already running.\n",
-      "\u001b[2K  Loading pipelines... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 83%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
-      "\u001b[?25h"
+      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-community-edition-linux...\n",
+      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240912171016.log.\n"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'dodgers'</span>.\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'dodgers'</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'dodgers'\u001b[0m\n"
+       "Connected to project \u001b[32m'dodgers'\u001b[0m.\n"
       ]
      },
      "metadata": {},
@@ -199,7 +185,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -218,7 +204,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -227,7 +213,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -333,7 +319,7 @@
        "[50400 rows x 2 columns]"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -351,7 +337,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -360,7 +346,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -370,7 +356,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -743,7 +729,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -754,7 +740,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -891,7 +877,7 @@
        "type: StringColumnView"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -919,7 +905,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -1220,7 +1206,7 @@
        "  0   data_full   50400   DataFrame"
       ]
      },
-     "execution_count": 12,
+     "execution_count": 13,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1263,7 +1249,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1284,7 +1270,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -1314,7 +1300,7 @@
        "         tags=['feature learning', 'fastprop'])"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1332,7 +1318,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
@@ -1352,25 +1338,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1391,7 +1364,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1400,7 +1373,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -1420,24 +1393,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1455,25 +1415,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Trying 526 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:08\u001b[0m5%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Trying 526 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:08\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1491,26 +1438,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:08.361847.\n",
+      "Time taken: 0:00:08.380775.\n",
       "\n",
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:08\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1521,31 +1455,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1554,7 +1475,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1567,7 +1488,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
     {
@@ -1587,24 +1508,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1622,24 +1530,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:09\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:10\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1657,7 +1552,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:09.975733.\n",
+      "Time taken: 0:00:10.075707.\n",
       "\n"
      ]
     },
@@ -1688,7 +1583,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1699,31 +1594,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:09\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1793,7 +1675,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:03:36</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:10:43</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1822,7 +1704,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:03:37</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:10:43</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1852,11 +1734,11 @@
       ],
       "text/plain": [
        "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-09-05 15:03:36   fastprop_train   y         5.4188    7.5347     0.699 \n",
-       "1   2024-09-05 15:03:37   fastprop_test    y         5.6151    7.8243     0.6747"
+       "0   2024-09-12 17:10:43   fastprop_train   y         5.4188    7.5347     0.699 \n",
+       "1   2024-09-12 17:10:43   fastprop_test    y         5.6151    7.8243     0.6747"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 22,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1874,7 +1756,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1884,7 +1766,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1897,7 +1779,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1914,7 +1796,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -1923,7 +1805,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:8m:58.96343\n",
+      "Time taken: 0h:8m:56.480767\n",
       "\n"
      ]
     }
@@ -1937,7 +1819,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1952,7 +1834,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1967,7 +1849,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
@@ -1997,7 +1879,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2014,7 +1896,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -2034,24 +1916,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2143,7 +2012,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2154,7 +2023,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
@@ -2174,23 +2043,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2217,41 +2073,15 @@
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2269,7 +2099,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:03.767338.\n",
+      "Time taken: 0:00:03.540051.\n",
       "\n"
      ]
     },
@@ -2300,7 +2130,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2311,31 +2141,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2405,7 +2222,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:15:36</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:22:38</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2434,7 +2251,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:15:36</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:22:38</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2464,11 +2281,11 @@
       ],
       "text/plain": [
        "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-09-05 15:15:36   featuretools_train   y         5.4482    7.568      0.6962\n",
-       "1   2024-09-05 15:15:36   featuretools_test    y         6.0863    8.5009     0.6498"
+       "0   2024-09-12 17:22:38   featuretools_train   y         5.4482    7.568      0.6962\n",
+       "1   2024-09-12 17:22:38   featuretools_test    y         6.0863    8.5009     0.6498"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2486,7 +2303,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2503,16 +2320,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:14<00:00,  2.83it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.78it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  5.93it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:12<00:00,  3.14it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.17it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.30it/s]\n"
      ]
     },
     {
@@ -2520,7 +2337,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 13 features...\n",
-      "Time taken: 0h:0m:34.625442\n",
+      "Time taken: 0h:0m:32.318403\n",
       "\n"
      ]
     },
@@ -2528,9 +2345,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.28it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 18.13it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 17.05it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.97it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 18.81it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 19.18it/s]\n"
      ]
     }
    ],
@@ -2543,7 +2360,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2558,7 +2375,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2569,7 +2386,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
     {
@@ -2599,7 +2416,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 37,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2614,7 +2431,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
@@ -2634,24 +2451,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2678,25 +2482,12 @@
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
@@ -2704,22 +2495,9 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K\u001b[32m⠧\u001b[0m XGBoost: Trained tree 96.  \u001b[31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[35m 96%\u001b[0m • \u001b[36m00:01\u001b[0m"
+      "\u001b[2K⠸ XGBoost: Trained tree 99.  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸  99% • 00:01"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2737,7 +2515,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:01.469758.\n",
+      "Time taken: 0:00:01.500237.\n",
       "\n"
      ]
     },
@@ -2768,7 +2546,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 37,
+     "execution_count": 38,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2779,31 +2557,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2873,7 +2638,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:16:24</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:23:23</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2902,7 +2667,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:16:24</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:23:23</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2932,11 +2697,11 @@
       ],
       "text/plain": [
        "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-09-05 15:16:24   tsfresh_train   y         6.3146    8.2348     0.6418\n",
-       "1   2024-09-05 15:16:24   tsfresh_test    y         6.7886    8.9134     0.5778"
+       "0   2024-09-12 17:23:23   tsfresh_train   y         6.3146    8.2348     0.6418\n",
+       "1   2024-09-12 17:23:23   tsfresh_test    y         6.7886    8.9134     0.5778"
       ]
      },
-     "execution_count": 38,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2954,7 +2719,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -3002,7 +2767,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [
     {
@@ -3039,9 +2804,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:11.918092</td>\n",
+       "      <td>0 days 00:00:12.036174</td>\n",
        "      <td>526</td>\n",
-       "      <td>44.134522</td>\n",
+       "      <td>43.702474</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.674740</td>\n",
@@ -3050,22 +2815,22 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:08:58.964287</td>\n",
+       "      <td>0 days 00:08:56.482944</td>\n",
        "      <td>59</td>\n",
-       "      <td>0.109469</td>\n",
-       "      <td>45.222363</td>\n",
-       "      <td>403.168329</td>\n",
+       "      <td>0.109976</td>\n",
+       "      <td>44.572548</td>\n",
+       "      <td>397.383577</td>\n",
        "      <td>0.649768</td>\n",
        "      <td>8.500887</td>\n",
        "      <td>6.086277</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:34.625580</td>\n",
+       "      <td>0 days 00:00:32.318518</td>\n",
        "      <td>12</td>\n",
-       "      <td>0.346565</td>\n",
-       "      <td>2.905296</td>\n",
-       "      <td>127.348619</td>\n",
+       "      <td>0.371304</td>\n",
+       "      <td>2.685116</td>\n",
+       "      <td>117.699939</td>\n",
        "      <td>0.577811</td>\n",
        "      <td>8.913408</td>\n",
        "      <td>6.788610</td>\n",
@@ -3076,14 +2841,14 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:11.918092           526            44.134522   \n",
-       "featuretools    0 days 00:08:58.964287            59             0.109469   \n",
-       "tsfresh         0 days 00:00:34.625580            12             0.346565   \n",
+       "getML: FastProp 0 days 00:00:12.036174           526            43.702474   \n",
+       "featuretools    0 days 00:08:56.482944            59             0.109976   \n",
+       "tsfresh         0 days 00:00:32.318518            12             0.371304   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.674740   \n",
-       "featuretools              45.222363                      403.168329  0.649768   \n",
-       "tsfresh                    2.905296                      127.348619  0.577811   \n",
+       "featuretools              44.572548                      397.383577  0.649768   \n",
+       "tsfresh                    2.685116                      117.699939  0.577811   \n",
        "\n",
        "                     rmse       mae  \n",
        "getML: FastProp  7.824273  5.615138  \n",
@@ -3091,7 +2856,7 @@
        "tsfresh          8.913408  6.788610  "
       ]
      },
-     "execution_count": 40,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3102,7 +2867,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [],
    "source": [
diff --git a/fastprop_benchmark/interstate94_prop.ipynb b/fastprop_benchmark/interstate94_prop.ipynb
index 7023907..1cd414c 100644
--- a/fastprop_benchmark/interstate94_prop.ipynb
+++ b/fastprop_benchmark/interstate94_prop.ipynb
@@ -119,30 +119,19 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML Engine is already running.\n"
+      "getML Engine is already running.\n",
+      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[?25h"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'interstate94'</span>.\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'interstate94'</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'interstate94'\u001b[0m\n"
+       "Connected to project \u001b[32m'interstate94'\u001b[0m.\n"
       ]
      },
      "metadata": {},
@@ -156,7 +145,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -167,7 +156,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -197,40 +186,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading traffic<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Loading traffic\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m1.2/1.2 MB\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
+      "\u001b[2K  Downloading traffic... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 1.2/1.2 MB • 00:00\n",
       "\u001b[?25h"
      ]
     }
@@ -241,7 +204,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -250,7 +213,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -974,7 +937,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -992,7 +955,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1001,7 +964,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -1310,7 +1273,7 @@
        "  0   traffic   24096   DataFrame"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1347,7 +1310,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1368,7 +1331,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -1398,7 +1361,7 @@
        "         tags=['feature learning', 'fastprop'])"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1416,7 +1379,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -1436,25 +1399,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1475,7 +1425,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1484,7 +1434,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -1504,24 +1454,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1539,25 +1476,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Trying 365 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0m2%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Trying 365 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1575,26 +1499,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:03.025341.\n",
+      "Time taken: 0:00:03.061732.\n",
       "\n",
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1605,31 +1516,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -1638,7 +1536,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1651,7 +1549,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -1671,24 +1569,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1706,24 +1591,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:05\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:05\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1741,7 +1613,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:05.318526.\n",
+      "Time taken: 0:00:05.329233.\n",
       "\n"
      ]
     },
@@ -1772,7 +1644,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1783,31 +1655,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:05\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1877,7 +1736,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:56:19</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:36:45</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1906,7 +1765,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 15:56:19</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:36:46</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1918,11 +1777,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">180.5073</td>\n",
+       "              <td class=\"float\">180.4867</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">261.9896</td>\n",
+       "              <td class=\"float\">261.9389</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1936,11 +1795,11 @@
       ],
       "text/plain": [
        "    date time             set used         target                 mae        rmse   rsquared\n",
-       "0   2024-09-05 15:56:19   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
-       "1   2024-09-05 15:56:19   fastprop_test    traffic_volume    180.5073    261.9896     0.9827"
+       "0   2024-09-12 16:36:45   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
+       "1   2024-09-12 16:36:46   fastprop_test    traffic_volume    180.4867    261.9389     0.9827"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 19,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1958,7 +1817,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1968,7 +1827,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1981,7 +1840,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1998,7 +1857,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {
     "tags": []
    },
@@ -2009,7 +1868,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:4m:30.436553\n",
+      "Time taken: 0h:4m:33.369809\n",
       "\n"
      ]
     }
@@ -2023,7 +1882,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2044,7 +1903,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2059,7 +1918,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -2089,7 +1948,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 26,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2106,7 +1965,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
@@ -2126,24 +1985,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2164,7 +2010,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
@@ -2184,23 +2030,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2218,24 +2051,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:02\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2253,7 +2073,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:02.084733.\n",
+      "Time taken: 0:00:02.103350.\n",
       "\n"
      ]
     },
@@ -2284,7 +2104,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2295,31 +2115,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:02\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2389,7 +2196,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 16:02:00</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:42:31</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2418,7 +2225,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 16:02:00</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:42:31</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2448,11 +2255,11 @@
       ],
       "text/plain": [
        "    date time             set used             target                 mae        rmse   rsquared\n",
-       "0   2024-09-05 16:02:00   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
-       "1   2024-09-05 16:02:00   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
+       "0   2024-09-12 16:42:31   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
+       "1   2024-09-12 16:42:31   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2479,7 +2286,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2520,7 +2327,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
@@ -2557,22 +2364,22 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:04.769646</td>\n",
+       "      <td>0 days 00:00:04.800454</td>\n",
        "      <td>461</td>\n",
-       "      <td>96.655712</td>\n",
-       "      <td>1.000000</td>\n",
+       "      <td>96.033804</td>\n",
+       "      <td>1.00000</td>\n",
        "      <td>1.000000</td>\n",
-       "      <td>0.982671</td>\n",
-       "      <td>261.989623</td>\n",
-       "      <td>180.507339</td>\n",
+       "      <td>0.982678</td>\n",
+       "      <td>261.938873</td>\n",
+       "      <td>180.486734</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:04:30.437045</td>\n",
+       "      <td>0 days 00:04:33.370925</td>\n",
        "      <td>59</td>\n",
-       "      <td>0.218165</td>\n",
-       "      <td>56.699605</td>\n",
-       "      <td>443.038759</td>\n",
+       "      <td>0.215824</td>\n",
+       "      <td>56.94689</td>\n",
+       "      <td>444.963603</td>\n",
        "      <td>0.974582</td>\n",
        "      <td>317.519976</td>\n",
        "      <td>210.198793</td>\n",
@@ -2583,19 +2390,19 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:04.769646           461            96.655712   \n",
-       "featuretools    0 days 00:04:30.437045            59             0.218165   \n",
+       "getML: FastProp 0 days 00:00:04.800454           461            96.033804   \n",
+       "featuretools    0 days 00:04:33.370925            59             0.215824   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
-       "getML: FastProp            1.000000                        1.000000  0.982671   \n",
-       "featuretools              56.699605                      443.038759  0.974582   \n",
+       "getML: FastProp             1.00000                        1.000000  0.982678   \n",
+       "featuretools               56.94689                      444.963603  0.974582   \n",
        "\n",
        "                       rmse         mae  \n",
-       "getML: FastProp  261.989623  180.507339  \n",
+       "getML: FastProp  261.938873  180.486734  \n",
        "featuretools     317.519976  210.198793  "
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2606,7 +2413,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [],
    "source": [
diff --git a/fastprop_benchmark/occupancy_prop.ipynb b/fastprop_benchmark/occupancy_prop.ipynb
index 32c6e6b..ea023dc 100644
--- a/fastprop_benchmark/occupancy_prop.ipynb
+++ b/fastprop_benchmark/occupancy_prop.ipynb
@@ -130,30 +130,19 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML Engine is already running.\n"
+      "getML Engine is already running.\n",
+      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[?25h"
      ]
     },
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'occupancy'</span>.\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'occupancy'</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'occupancy'\u001b[0m\n"
+       "Connected to project \u001b[32m'occupancy'\u001b[0m.\n"
       ]
      },
      "metadata": {},
@@ -167,7 +156,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -178,7 +167,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -206,119 +195,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_train<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Loading population_train\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m554.6/554.6 kB\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_test<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Loading population_test\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m668.3/668.3 kB\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Loading population_validation<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Loading population_validation\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m186.5/186.5 kB\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "data_test, data_train, data_validate = getml.datasets.load_occupancy(roles=True)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -339,7 +225,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {
     "tags": [
      "hide_input"
@@ -1491,7 +1377,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1525,7 +1411,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -1856,7 +1742,7 @@
        "  0   data_all   20560   DataFrame"
       ]
      },
-     "execution_count": 8,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1891,7 +1777,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1913,7 +1799,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1926,7 +1812,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -1946,24 +1832,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -1991,7 +1864,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2000,7 +1873,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [
     {
@@ -2020,23 +1893,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2054,24 +1914,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Trying 331 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0m0%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2089,26 +1936,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:01.035566.\n",
+      "Time taken: 0:00:01.047819.\n",
       "\n",
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2119,31 +1953,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m[0m • \u001b[36m--:--\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -2160,7 +1981,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2173,7 +1994,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
@@ -2193,24 +2014,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2241,23 +2049,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2275,24 +2070,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2310,7 +2092,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.948174.\n",
+      "Time taken: 0:00:04.853148.\n",
       "\n"
      ]
     },
@@ -2341,7 +2123,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2354,31 +2136,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2448,7 +2217,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 17:46:11</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:20:28</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2477,7 +2246,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 17:46:12</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:20:29</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2497,7 +2266,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.044287</td>\n",
+       "              <td class=\"float\">0.044213</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -2507,11 +2276,11 @@
       ],
       "text/plain": [
        "    date time             set used         target      accuracy       auc   cross entropy\n",
-       "0   2024-09-05 17:46:11   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
-       "1   2024-09-05 17:46:12   fastprop_test    Occupancy     0.9888    0.9982        0.044287"
+       "0   2024-09-12 16:20:28   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
+       "1   2024-09-12 16:20:29   fastprop_test    Occupancy     0.9888    0.9982        0.044213"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2529,7 +2298,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2539,7 +2308,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2552,7 +2321,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -2576,7 +2345,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
@@ -2585,7 +2354,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 262 features...\n",
-      "Time taken: 0h:7m:18.06737\n",
+      "Time taken: 0h:7m:28.128247\n",
       "\n"
      ]
     }
@@ -2614,7 +2383,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
@@ -2644,7 +2413,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 23,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2661,7 +2430,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [
     {
@@ -2681,24 +2450,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2790,7 +2546,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 23,
+     "execution_count": 24,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2801,7 +2557,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -2821,23 +2577,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2864,41 +2607,15 @@
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -2916,7 +2633,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:03.188635.\n",
+      "Time taken: 0:00:03.326992.\n",
       "\n"
      ]
     },
@@ -2947,7 +2664,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2958,31 +2675,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3052,7 +2756,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 17:59:45</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:34:15</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3081,7 +2785,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 17:59:45</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:34:16</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3111,11 +2815,11 @@
       ],
       "text/plain": [
        "    date time             set used             target      accuracy       auc   cross entropy\n",
-       "0   2024-09-05 17:59:45   featuretools_train   Occupancy     0.9995    1.            0.005065\n",
-       "1   2024-09-05 17:59:45   featuretools_test    Occupancy     0.9885    0.9972        0.049236"
+       "0   2024-09-12 16:34:15   featuretools_train   Occupancy     0.9995    1.            0.005065\n",
+       "1   2024-09-12 16:34:16   featuretools_test    Occupancy     0.9885    0.9972        0.049236"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 26,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3133,7 +2837,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {
     "lines_to_next_cell": 2
    },
@@ -3142,9 +2846,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.45it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.26it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.59it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.60it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.46it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.32it/s]\n"
      ]
     },
     {
@@ -3152,7 +2856,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 65 features...\n",
-      "Time taken: 0h:0m:16.408325\n",
+      "Time taken: 0h:0m:17.070296\n",
       "\n"
      ]
     },
@@ -3160,9 +2864,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 19.16it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.80it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:03<00:00, 10.21it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 19.76it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.89it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.92it/s]\n"
      ]
     }
    ],
@@ -3190,7 +2894,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
@@ -3220,7 +2924,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3235,7 +2939,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
@@ -3255,24 +2959,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3364,7 +3055,7 @@
        "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 28,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3375,7 +3066,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -3395,23 +3086,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3438,41 +3116,15 @@
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:01\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3490,7 +3142,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:01.717708.\n",
+      "Time taken: 0:00:01.784330.\n",
       "\n"
      ]
     },
@@ -3521,7 +3173,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3532,31 +3184,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:01\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -3626,7 +3265,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:00:17</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:34:49</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3655,7 +3294,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:00:17</td>\n",
+       "              <td class=\"datetime\">2024-09-12 16:34:49</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -3685,11 +3324,11 @@
       ],
       "text/plain": [
        "    date time             set used        target      accuracy       auc   cross entropy\n",
-       "0   2024-09-05 18:00:17   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
-       "1   2024-09-05 18:00:17   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
+       "0   2024-09-12 16:34:49   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
+       "1   2024-09-12 16:34:49   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3709,7 +3348,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {
     "lines_to_next_cell": 2
    },
@@ -3763,7 +3402,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {
     "tags": []
    },
@@ -3802,33 +3441,33 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:01.816878</td>\n",
+       "      <td>0 days 00:00:01.854957</td>\n",
        "      <td>289</td>\n",
-       "      <td>159.058374</td>\n",
+       "      <td>155.787506</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.988823</td>\n",
-       "      <td>0.998173</td>\n",
-       "      <td>0.044287</td>\n",
+       "      <td>0.998166</td>\n",
+       "      <td>0.044213</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:07:18.068960</td>\n",
+       "      <td>0 days 00:07:28.129456</td>\n",
        "      <td>103</td>\n",
-       "      <td>0.235123</td>\n",
-       "      <td>241.110829</td>\n",
-       "      <td>676.490695</td>\n",
+       "      <td>0.229844</td>\n",
+       "      <td>241.584822</td>\n",
+       "      <td>677.795763</td>\n",
        "      <td>0.988455</td>\n",
        "      <td>0.997207</td>\n",
        "      <td>0.049236</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:16.408459</td>\n",
+       "      <td>0 days 00:00:17.070465</td>\n",
        "      <td>60</td>\n",
-       "      <td>3.656655</td>\n",
-       "      <td>9.031129</td>\n",
-       "      <td>43.498330</td>\n",
+       "      <td>3.514840</td>\n",
+       "      <td>9.202620</td>\n",
+       "      <td>44.322792</td>\n",
        "      <td>0.987718</td>\n",
        "      <td>0.997861</td>\n",
        "      <td>0.049359</td>\n",
@@ -3839,22 +3478,22 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:01.816878           289           159.058374   \n",
-       "featuretools    0 days 00:07:18.068960           103             0.235123   \n",
-       "tsfresh         0 days 00:00:16.408459            60             3.656655   \n",
+       "getML: FastProp 0 days 00:00:01.854957           289           155.787506   \n",
+       "featuretools    0 days 00:07:28.129456           103             0.229844   \n",
+       "tsfresh         0 days 00:00:17.070465            60             3.514840   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  accuracy  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.988823   \n",
-       "featuretools             241.110829                      676.490695  0.988455   \n",
-       "tsfresh                    9.031129                       43.498330  0.987718   \n",
+       "featuretools             241.584822                      677.795763  0.988455   \n",
+       "tsfresh                    9.202620                       44.322792  0.987718   \n",
        "\n",
        "                      auc  cross_entropy  \n",
-       "getML: FastProp  0.998173       0.044287  \n",
+       "getML: FastProp  0.998166       0.044213  \n",
        "featuretools     0.997207       0.049236  \n",
        "tsfresh          0.997861       0.049359  "
       ]
      },
-     "execution_count": 32,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3865,7 +3504,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [],
    "source": [
diff --git a/fastprop_benchmark/robot_prop.ipynb b/fastprop_benchmark/robot_prop.ipynb
index 8b25d87..3f29ccb 100644
--- a/fastprop_benchmark/robot_prop.ipynb
+++ b/fastprop_benchmark/robot_prop.ipynb
@@ -128,24 +128,11 @@
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'robot'</span>.\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'robot'</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'robot'\u001b[0m\n"
+       "Connected to project \u001b[32m'robot'\u001b[0m.\n"
       ]
      },
      "metadata": {},
@@ -159,7 +146,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -170,7 +157,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -198,7 +185,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -220,7 +207,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Downloading... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[32m14.7/14.7 MB\u001b[0m • \u001b[33m00:00\u001b[0m4.7 MB\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Downloading robot-demo.csv... ━━━━━━━━━━━━━━━━━━━━ 100% • 14.7/14.7 MB • 00:00\n",
       "\u001b[?25h"
      ]
     }
@@ -233,7 +220,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -14896,7 +14883,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -14918,7 +14905,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -14937,7 +14924,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -29600,7 +29587,7 @@
        "type: getml.DataFrame"
       ]
      },
-     "execution_count": 8,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -29620,7 +29607,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -29757,7 +29744,7 @@
        "type: StringColumnView"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -29769,7 +29756,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -30077,7 +30064,7 @@
        "  0   data_all   15001   View"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30105,7 +30092,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30116,7 +30103,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30129,7 +30116,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [
     {
@@ -30149,24 +30136,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m5m 50%\u001b[0m • \u001b[36m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30187,7 +30161,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30196,7 +30170,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
@@ -30216,23 +30190,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30250,24 +30211,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Trying 134 features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Trying 134 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30285,26 +30233,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:00.037119.\n",
+      "Time taken: 0:00:00.043565.\n",
       "\n",
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -30315,31 +30250,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  FastProp: Building features... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
     }
    ],
    "source": [
@@ -30348,7 +30270,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30361,7 +30283,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
@@ -30381,24 +30303,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30477,7 +30386,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_126&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_121&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30494,7 +30403,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_127&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_122&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30511,7 +30420,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_129&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_123&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30528,7 +30437,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_130&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_127&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30545,7 +30454,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_132&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_129&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30562,7 +30471,7 @@
        "4   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
       ]
      },
-     "execution_count": 18,
+     "execution_count": 19,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30573,7 +30482,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
     {
@@ -30593,23 +30502,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30636,41 +30532,15 @@
      "metadata": {},
      "output_type": "display_data"
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30688,7 +30558,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.133426.\n",
+      "Time taken: 0:00:04.089082.\n",
       "\n"
      ]
     },
@@ -30719,7 +30589,7 @@
        "         tags=['prediction', 'fastprop'])"
       ]
      },
-     "execution_count": 19,
+     "execution_count": 20,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30730,31 +30600,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -30824,7 +30681,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:17:16</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:26:03</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30853,7 +30710,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:17:17</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:26:03</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30865,11 +30722,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.5515</td>\n",
+       "              <td class=\"float\">0.5516</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.7236</td>\n",
+       "              <td class=\"float\">0.7237</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30883,11 +30740,11 @@
       ],
       "text/plain": [
        "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-09-05 18:17:16   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
-       "1   2024-09-05 18:17:17   fastprop_test    f_x       0.5515    0.7236     0.9951"
+       "0   2024-09-12 17:26:03   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
+       "1   2024-09-12 17:26:03   fastprop_test    f_x       0.5516    0.7237     0.9951"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -30905,7 +30762,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30915,7 +30772,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -30936,7 +30793,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [
     {
@@ -31187,7 +31044,7 @@
        "[10500 rows x 13 columns]"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 24,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31198,7 +31055,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31214,7 +31071,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -31223,7 +31080,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 442 features...\n",
-      "Time taken: 0h:14m:35.631365\n",
+      "Time taken: 0h:14m:36.192123\n",
       "\n"
      ]
     }
@@ -31237,7 +31094,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
@@ -31325,10 +31182,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
        "      <td>-11.0300</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:00</td>\n",
@@ -31349,10 +31206,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
        "      <td>-10.8480</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:01</td>\n",
@@ -31373,10 +31230,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
        "      <td>-10.6660</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:02</td>\n",
@@ -31397,10 +31254,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
        "      <td>-10.5070</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:03</td>\n",
@@ -31421,10 +31278,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
-       "      <td>1.725560e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726162e+09</td>\n",
        "      <td>-10.4130</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:04</td>\n",
@@ -31469,10 +31326,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
        "      <td>-9.7673</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:55</td>\n",
@@ -31493,10 +31350,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
        "      <td>-9.9200</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:56</td>\n",
@@ -31517,10 +31374,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
        "      <td>-9.7743</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:57</td>\n",
@@ -31541,10 +31398,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
        "      <td>-8.6109</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:58</td>\n",
@@ -31565,10 +31422,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
-       "      <td>1.725550e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726151e+09</td>\n",
        "      <td>-8.4345</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:59</td>\n",
@@ -31693,59 +31550,59 @@
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 4)  \\\n",
        "_featuretools_index                                          \n",
-       "0                                             1.725560e+09   \n",
-       "1                                             1.725560e+09   \n",
-       "2                                             1.725560e+09   \n",
-       "3                                             1.725560e+09   \n",
-       "4                                             1.725560e+09   \n",
+       "0                                             1.726162e+09   \n",
+       "1                                             1.726162e+09   \n",
+       "2                                             1.726162e+09   \n",
+       "3                                             1.726162e+09   \n",
+       "4                                             1.726162e+09   \n",
        "...                                                    ...   \n",
-       "10495                                         1.725550e+09   \n",
-       "10496                                         1.725550e+09   \n",
-       "10497                                         1.725550e+09   \n",
-       "10498                                         1.725550e+09   \n",
-       "10499                                         1.725550e+09   \n",
+       "10495                                         1.726151e+09   \n",
+       "10496                                         1.726151e+09   \n",
+       "10497                                         1.726151e+09   \n",
+       "10498                                         1.726151e+09   \n",
+       "10499                                         1.726151e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 30)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.725560e+09   \n",
-       "1                                              1.725560e+09   \n",
-       "2                                              1.725560e+09   \n",
-       "3                                              1.725560e+09   \n",
-       "4                                              1.725560e+09   \n",
+       "0                                              1.726162e+09   \n",
+       "1                                              1.726162e+09   \n",
+       "2                                              1.726162e+09   \n",
+       "3                                              1.726162e+09   \n",
+       "4                                              1.726162e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.725550e+09   \n",
-       "10496                                          1.725550e+09   \n",
-       "10497                                          1.725550e+09   \n",
-       "10498                                          1.725550e+09   \n",
-       "10499                                          1.725550e+09   \n",
+       "10495                                          1.726151e+09   \n",
+       "10496                                          1.726151e+09   \n",
+       "10497                                          1.726151e+09   \n",
+       "10498                                          1.726151e+09   \n",
+       "10499                                          1.726151e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 77)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.725560e+09   \n",
-       "1                                              1.725560e+09   \n",
-       "2                                              1.725560e+09   \n",
-       "3                                              1.725560e+09   \n",
-       "4                                              1.725560e+09   \n",
+       "0                                              1.726162e+09   \n",
+       "1                                              1.726162e+09   \n",
+       "2                                              1.726162e+09   \n",
+       "3                                              1.726162e+09   \n",
+       "4                                              1.726162e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.725550e+09   \n",
-       "10496                                          1.725550e+09   \n",
-       "10497                                          1.725550e+09   \n",
-       "10498                                          1.725550e+09   \n",
-       "10499                                          1.725550e+09   \n",
+       "10495                                          1.726151e+09   \n",
+       "10496                                          1.726151e+09   \n",
+       "10497                                          1.726151e+09   \n",
+       "10498                                          1.726151e+09   \n",
+       "10499                                          1.726151e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 7)      f_x  id  \\\n",
        "_featuretools_index                                                       \n",
-       "0                                             1.725560e+09 -11.0300   1   \n",
-       "1                                             1.725560e+09 -10.8480   1   \n",
-       "2                                             1.725560e+09 -10.6660   1   \n",
-       "3                                             1.725560e+09 -10.5070   1   \n",
-       "4                                             1.725560e+09 -10.4130   1   \n",
+       "0                                             1.726162e+09 -11.0300   1   \n",
+       "1                                             1.726162e+09 -10.8480   1   \n",
+       "2                                             1.726162e+09 -10.6660   1   \n",
+       "3                                             1.726162e+09 -10.5070   1   \n",
+       "4                                             1.726162e+09 -10.4130   1   \n",
        "...                                                    ...      ...  ..   \n",
-       "10495                                         1.725550e+09  -9.7673   1   \n",
-       "10496                                         1.725550e+09  -9.9200   1   \n",
-       "10497                                         1.725550e+09  -9.7743   1   \n",
-       "10498                                         1.725550e+09  -8.6109   1   \n",
-       "10499                                         1.725550e+09  -8.4345   1   \n",
+       "10495                                         1.726151e+09  -9.7673   1   \n",
+       "10496                                         1.726151e+09  -9.9200   1   \n",
+       "10497                                         1.726151e+09  -9.7743   1   \n",
+       "10498                                         1.726151e+09  -8.6109   1   \n",
+       "10499                                         1.726151e+09  -8.4345   1   \n",
        "\n",
        "                                     ds  \n",
        "_featuretools_index                      \n",
@@ -31764,7 +31621,7 @@
        "[10500 rows x 203 columns]"
       ]
      },
-     "execution_count": 26,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31775,7 +31632,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31796,7 +31653,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31811,7 +31668,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -31841,7 +31698,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31858,7 +31715,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
@@ -31878,24 +31735,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -31913,24 +31757,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:04\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -31948,7 +31779,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.440980.\n",
+      "Time taken: 0:00:04.582992.\n",
       "\n"
      ]
     },
@@ -31979,7 +31810,7 @@
        "         tags=['prediction', 'featuretools'])"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -31990,31 +31821,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:04\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -32084,7 +31902,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:38:13</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:47:07</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32096,11 +31914,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.4395</td>\n",
+       "              <td class=\"float\">0.4394</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.5835</td>\n",
+       "              <td class=\"float\">0.5831</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32113,7 +31931,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:38:13</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:47:07</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32125,11 +31943,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.572</td>\n",
+       "              <td class=\"float\">0.571</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.7486</td>\n",
+       "              <td class=\"float\">0.7484</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32143,11 +31961,11 @@
       ],
       "text/plain": [
        "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-09-05 18:38:13   featuretools_train   f_x       0.4395    0.5835     0.9962\n",
-       "1   2024-09-05 18:38:13   featuretools_test    f_x       0.572     0.7486     0.9948"
+       "0   2024-09-12 17:47:07   featuretools_train   f_x       0.4394    0.5831     0.9962\n",
+       "1   2024-09-12 17:47:07   featuretools_test    f_x       0.571     0.7484     0.9948"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32165,7 +31983,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32180,16 +31998,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.96it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:10<00:00,  3.95it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:10<00:00,  3.94it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:04<00:00,  9.82it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:12<00:00,  3.14it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:11<00:00,  3.60it/s]\n"
      ]
     },
     {
@@ -32197,7 +32015,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 130 features...\n",
-      "Time taken: 0h:0m:26.245796\n",
+      "Time taken: 0h:0m:31.641678\n",
       "\n"
      ]
     },
@@ -32205,9 +32023,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 35.68it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.43it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.14it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 28.10it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.14it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.33it/s]\n"
      ]
     }
    ],
@@ -32220,7 +32038,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32241,7 +32059,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32252,7 +32070,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
     {
@@ -32282,7 +32100,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 37,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32297,7 +32115,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
@@ -32317,24 +32135,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m--:--\u001b[0m\n",
-      "\u001b[2K  Checking... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -32355,7 +32160,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
@@ -32375,23 +32180,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -32409,24 +32201,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
-      "\u001b[2K  XGBoost: Training as predictor... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:03\u001b[0mm • \u001b[36m00:01\u001b[0m00:01\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -32444,7 +32223,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:03.945516.\n",
+      "Time taken: 0:00:04.062273.\n",
       "\n"
      ]
     },
@@ -32475,7 +32254,7 @@
        "         tags=['predicition', 'tsfresh'])"
       ]
      },
-     "execution_count": 38,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32486,31 +32265,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[2K  Staging... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:03\u001b[0m\n",
-      "\u001b[2K  Preprocessing... \u001b[32m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m100%\u001b[0m • \u001b[33m00:00\u001b[0m00:00\u001b[0m\n",
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[?25h"
      ]
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "data": {
       "text/html": [
@@ -32580,7 +32346,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:38:57</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:47:59</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32609,7 +32375,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-05 18:38:57</td>\n",
+       "              <td class=\"datetime\">2024-09-12 17:47:59</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32639,11 +32405,11 @@
       ],
       "text/plain": [
        "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-09-05 18:38:57   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
-       "1   2024-09-05 18:38:57   tsfresh_test    f_x       0.5986    0.7906     0.9938"
+       "0   2024-09-12 17:47:59   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
+       "1   2024-09-12 17:47:59   tsfresh_test    f_x       0.5986    0.7906     0.9938"
       ]
      },
-     "execution_count": 39,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32661,7 +32427,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -32709,7 +32475,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [
     {
@@ -32746,33 +32512,33 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:00.424167</td>\n",
+       "      <td>0 days 00:00:00.434939</td>\n",
        "      <td>134</td>\n",
-       "      <td>315.955766</td>\n",
+       "      <td>308.071473</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
-       "      <td>0.995059</td>\n",
-       "      <td>0.723622</td>\n",
-       "      <td>0.551521</td>\n",
+       "      <td>0.995058</td>\n",
+       "      <td>0.723674</td>\n",
+       "      <td>0.551593</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:14:35.632221</td>\n",
+       "      <td>0 days 00:14:36.193549</td>\n",
        "      <td>158</td>\n",
-       "      <td>0.180441</td>\n",
-       "      <td>2064.357248</td>\n",
-       "      <td>1751.019273</td>\n",
-       "      <td>0.994803</td>\n",
-       "      <td>0.748614</td>\n",
-       "      <td>0.572019</td>\n",
+       "      <td>0.180325</td>\n",
+       "      <td>2014.520540</td>\n",
+       "      <td>1708.419285</td>\n",
+       "      <td>0.994832</td>\n",
+       "      <td>0.748423</td>\n",
+       "      <td>0.571029</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:26.245968</td>\n",
+       "      <td>0 days 00:00:31.641862</td>\n",
        "      <td>120</td>\n",
-       "      <td>4.572139</td>\n",
-       "      <td>61.876497</td>\n",
-       "      <td>69.104581</td>\n",
+       "      <td>3.792447</td>\n",
+       "      <td>72.750114</td>\n",
+       "      <td>81.232902</td>\n",
        "      <td>0.993836</td>\n",
        "      <td>0.790602</td>\n",
        "      <td>0.598600</td>\n",
@@ -32783,22 +32549,22 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:00.424167           134           315.955766   \n",
-       "featuretools    0 days 00:14:35.632221           158             0.180441   \n",
-       "tsfresh         0 days 00:00:26.245968           120             4.572139   \n",
+       "getML: FastProp 0 days 00:00:00.434939           134           308.071473   \n",
+       "featuretools    0 days 00:14:36.193549           158             0.180325   \n",
+       "tsfresh         0 days 00:00:31.641862           120             3.792447   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
-       "getML: FastProp            1.000000                        1.000000  0.995059   \n",
-       "featuretools            2064.357248                     1751.019273  0.994803   \n",
-       "tsfresh                   61.876497                       69.104581  0.993836   \n",
+       "getML: FastProp            1.000000                        1.000000  0.995058   \n",
+       "featuretools            2014.520540                     1708.419285  0.994832   \n",
+       "tsfresh                   72.750114                       81.232902  0.993836   \n",
        "\n",
        "                     rmse       mae  \n",
-       "getML: FastProp  0.723622  0.551521  \n",
-       "featuretools     0.748614  0.572019  \n",
+       "getML: FastProp  0.723674  0.551593  \n",
+       "featuretools     0.748423  0.571029  \n",
        "tsfresh          0.790602  0.598600  "
       ]
      },
-     "execution_count": 41,
+     "execution_count": 42,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -32809,7 +32575,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 43,
    "metadata": {},
    "outputs": [],
    "source": [

From 699b3b4f86015cbb2ae08f87953812b0b2aa9111 Mon Sep 17 00:00:00 2001
From: Alexandros Ladas <alexandros@getml.com>
Date: Fri, 13 Sep 2024 14:44:21 +0200
Subject: [PATCH 3/3] Updated dependencies tsfresh + featuretools

---
 fastprop_benchmark/air_pollution_prop.ipynb   | 8722 ++++++++--------
 .../comparisons/air_pollution.csv             |    6 +-
 fastprop_benchmark/comparisons/dodgers.csv    |    6 +-
 .../comparisons/interstate94.csv              |    4 +-
 fastprop_benchmark/comparisons/occupancy.csv  |    6 +-
 fastprop_benchmark/comparisons/robot.csv      |    6 +-
 fastprop_benchmark/dodgers_prop.ipynb         | 5827 ++++++-----
 fastprop_benchmark/interstate94_prop.ipynb    |   63 +-
 fastprop_benchmark/occupancy_prop.ipynb       | 7112 ++++++-------
 fastprop_benchmark/robot_prop.ipynb           |  272 +-
 kaggle_notebooks/epilepsy_recognition.ipynb   | 9120 ++++++++---------
 requirements.txt                              |    4 +-
 12 files changed, 15598 insertions(+), 15550 deletions(-)

diff --git a/fastprop_benchmark/air_pollution_prop.ipynb b/fastprop_benchmark/air_pollution_prop.ipynb
index 59b0f7e..8ee4008 100644
--- a/fastprop_benchmark/air_pollution_prop.ipynb
+++ b/fastprop_benchmark/air_pollution_prop.ipynb
@@ -1,4348 +1,4380 @@
 {
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "tags": [
-     "hide_input"
-    ]
-   },
-   "source": [
-    "# Propositionalization: Predicting air pollution in Beijing\n",
-    "\n",
-    "In this notebook we will compare getML to featuretools and tsfresh, both of which open-source libraries for feature engineering. We find that advanced algorithms featured in getML yield significantly better predictions on this dataset. We then discuss why that is.\n",
-    "\n",
-    "Summary:\n",
-    "\n",
-    "- Prediction type: __Regression model__\n",
-    "- Domain: __Air pollution__\n",
-    "- Prediction target: __pm 2.5 concentration__ \n",
-    "- Source data: __Multivariate time series__\n",
-    "- Population size: __41757__"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "tags": [
-     "remove_cell_on_docs"
-    ]
-   },
-   "source": [
-    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/air_pollution_prop.ipynb\">\n",
-    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
-    "</a>"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Background\n",
-    "\n",
-    "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
-    "\n",
-    "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
-    "\n",
-    "As our example dataset, we use a publicly available dataset on air pollution in Beijing, China (https://archive.ics.uci.edu/dataset/381/beijing+pm2+5+data). For further details about the data set refer to [the full notebook](../air_pollution.ipynb)."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Analysis"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Table of contents\n",
-    "\n",
-    "1. [Loading data](#1.-Loading-data)\n",
-    "2. [Predictive modeling](#2.-Predictive-modeling)\n",
-    "3. [Comparison](#3.-Comparison)\n",
-    "4. [Conclusion](#4.-Conclusion)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Note: you may need to restart the kernel to use updated packages.\n"
-     ]
-    }
-   ],
-   "source": [
-    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "getML API version: 1.5.0\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "import os\n",
-    "import sys\n",
-    "\n",
-    "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
-    "from pathlib import Path\n",
-    "from urllib import request\n",
-    "import getml\n",
-    "import pandas as pd\n",
-    "\n",
-    "print(f\"getML API version: {getml.__version__}\\n\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
-    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
-    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "parent = Path(os.getcwd()).parent.as_posix()\n",
-    "\n",
-    "if parent not in sys.path:\n",
-    "    sys.path.append(parent)\n",
-    "\n",
-    "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 1. Loading data"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 1.1 Download from source\n",
-    "\n",
-    "We begin by downloading the data from the UCI Machine Learning repository."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "FEATURETOOLS_FILES = [\"featuretools_training.csv\", \"featuretools_test.csv\"]\n",
-    "\n",
-    "for fname in FEATURETOOLS_FILES:\n",
-    "    if not os.path.exists(fname):\n",
-    "        fname, res = request.urlretrieve(\n",
-    "            \"https://static.getml.com/datasets/air_pollution/featuretools/\" + fname,\n",
-    "            fname,\n",
-    "        )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "TSFRESH_FILES = [\"tsfresh_training.csv\", \"tsfresh_test.csv\"]\n",
-    "\n",
-    "for fname in TSFRESH_FILES:\n",
-    "    if not os.path.exists(fname):\n",
-    "        fname, res = request.urlretrieve(\n",
-    "            \"https://static.getml.com/datasets/air_pollution/tsfresh/\" + fname, fname\n",
-    "        )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "fname = \"PRSA_data_2010.1.1-2014.12.31.csv\"\n",
-    "\n",
-    "if not os.path.exists(fname):\n",
-    "    fname, res = request.urlretrieve(\n",
-    "        \"https://archive.ics.uci.edu/ml/machine-learning-databases/00381/\" + fname,\n",
-    "        fname,\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 1.2 Prepare data for tsfresh and getML\n",
-    "\n",
-    "Our our goal is to predict the pm2.5 concentration from factors such as weather or time of day. However, there are some **missing entries** for pm2.5, so we get rid of them."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_full_pandas = pd.read_csv(fname)\n",
-    "\n",
-    "data_full_pandas = data_full_pandas[\n",
-    "    data_full_pandas[\"pm2.5\"] == data_full_pandas[\"pm2.5\"]\n",
-    "]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "tsfresh requires a date column, so we build one."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def add_leading_zero(val):\n",
-    "    if len(str(val)) == 1:\n",
-    "        return \"0\" + str(val)\n",
-    "    return str(val)\n",
-    "\n",
-    "\n",
-    "data_full_pandas[\"month\"] = [add_leading_zero(val) for val in data_full_pandas[\"month\"]]\n",
-    "\n",
-    "data_full_pandas[\"day\"] = [add_leading_zero(val) for val in data_full_pandas[\"day\"]]\n",
-    "\n",
-    "data_full_pandas[\"hour\"] = [add_leading_zero(val) for val in data_full_pandas[\"hour\"]]\n",
-    "\n",
-    "\n",
-    "def make_date(year, month, day, hour):\n",
-    "    return year + \"-\" + month + \"-\" + day + \" \" + hour + \":00:00\"\n",
-    "\n",
-    "\n",
-    "data_full_pandas[\"date\"] = [\n",
-    "    make_date(str(year), month, day, hour)\n",
-    "    for year, month, day, hour in zip(\n",
-    "        data_full_pandas[\"year\"],\n",
-    "        data_full_pandas[\"month\"],\n",
-    "        data_full_pandas[\"day\"],\n",
-    "        data_full_pandas[\"hour\"],\n",
-    "    )\n",
-    "]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "tsfresh also requires the time series to have ids. Since there is only a single time series, that series has the same id."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_full_pandas[\"id\"] = 1"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The dataset now contains many columns that we do not need or that tsfresh cannot process. For instance, *cbwd* might actually contain useful information, but it is a categorical variable, which is difficult to handle for tsfresh, so we remove it."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We also want to split our data into a training and testing set."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_train_pandas = data_full_pandas[data_full_pandas[\"year\"] < 2014]\n",
-    "data_test_pandas = data_full_pandas[data_full_pandas[\"year\"] == 2014]\n",
-    "data_full_pandas = data_full_pandas"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def remove_unwanted_columns(df):\n",
-    "    del df[\"cbwd\"]\n",
-    "    del df[\"year\"]\n",
-    "    del df[\"month\"]\n",
-    "    del df[\"day\"]\n",
-    "    del df[\"hour\"]\n",
-    "    del df[\"No\"]\n",
-    "    return df\n",
-    "\n",
-    "\n",
-    "data_full_pandas = remove_unwanted_columns(data_full_pandas)\n",
-    "data_train_pandas = remove_unwanted_columns(data_train_pandas)\n",
-    "data_test_pandas = remove_unwanted_columns(data_test_pandas)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>pm2.5</th>\n",
-       "      <th>DEWP</th>\n",
-       "      <th>TEMP</th>\n",
-       "      <th>PRES</th>\n",
-       "      <th>Iws</th>\n",
-       "      <th>Is</th>\n",
-       "      <th>Ir</th>\n",
-       "      <th>date</th>\n",
-       "      <th>id</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>24</th>\n",
-       "      <td>129.0</td>\n",
-       "      <td>-16</td>\n",
-       "      <td>-4.0</td>\n",
-       "      <td>1020.0</td>\n",
-       "      <td>1.79</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 00:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25</th>\n",
-       "      <td>148.0</td>\n",
-       "      <td>-15</td>\n",
-       "      <td>-4.0</td>\n",
-       "      <td>1020.0</td>\n",
-       "      <td>2.68</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 01:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>26</th>\n",
-       "      <td>159.0</td>\n",
-       "      <td>-11</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1021.0</td>\n",
-       "      <td>3.57</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 02:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>27</th>\n",
-       "      <td>181.0</td>\n",
-       "      <td>-7</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1022.0</td>\n",
-       "      <td>5.36</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 03:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>28</th>\n",
-       "      <td>138.0</td>\n",
-       "      <td>-7</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1022.0</td>\n",
-       "      <td>6.25</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 04:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43819</th>\n",
-       "      <td>8.0</td>\n",
-       "      <td>-23</td>\n",
-       "      <td>-2.0</td>\n",
-       "      <td>1034.0</td>\n",
-       "      <td>231.97</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2014-12-31 19:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43820</th>\n",
-       "      <td>10.0</td>\n",
-       "      <td>-22</td>\n",
-       "      <td>-3.0</td>\n",
-       "      <td>1034.0</td>\n",
-       "      <td>237.78</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2014-12-31 20:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43821</th>\n",
-       "      <td>10.0</td>\n",
-       "      <td>-22</td>\n",
-       "      <td>-3.0</td>\n",
-       "      <td>1034.0</td>\n",
-       "      <td>242.70</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2014-12-31 21:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43822</th>\n",
-       "      <td>8.0</td>\n",
-       "      <td>-22</td>\n",
-       "      <td>-4.0</td>\n",
-       "      <td>1034.0</td>\n",
-       "      <td>246.72</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2014-12-31 22:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>43823</th>\n",
-       "      <td>12.0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>-3.0</td>\n",
-       "      <td>1034.0</td>\n",
-       "      <td>249.85</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2014-12-31 23:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>41757 rows × 9 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       pm2.5  DEWP  TEMP    PRES     Iws  Is  Ir                 date  id\n",
-       "24     129.0   -16  -4.0  1020.0    1.79   0   0  2010-01-02 00:00:00   1\n",
-       "25     148.0   -15  -4.0  1020.0    2.68   0   0  2010-01-02 01:00:00   1\n",
-       "26     159.0   -11  -5.0  1021.0    3.57   0   0  2010-01-02 02:00:00   1\n",
-       "27     181.0    -7  -5.0  1022.0    5.36   1   0  2010-01-02 03:00:00   1\n",
-       "28     138.0    -7  -5.0  1022.0    6.25   2   0  2010-01-02 04:00:00   1\n",
-       "...      ...   ...   ...     ...     ...  ..  ..                  ...  ..\n",
-       "43819    8.0   -23  -2.0  1034.0  231.97   0   0  2014-12-31 19:00:00   1\n",
-       "43820   10.0   -22  -3.0  1034.0  237.78   0   0  2014-12-31 20:00:00   1\n",
-       "43821   10.0   -22  -3.0  1034.0  242.70   0   0  2014-12-31 21:00:00   1\n",
-       "43822    8.0   -22  -4.0  1034.0  246.72   0   0  2014-12-31 22:00:00   1\n",
-       "43823   12.0   -21  -3.0  1034.0  249.85   0   0  2014-12-31 23:00:00   1\n",
-       "\n",
-       "[41757 rows x 9 columns]"
-      ]
-     },
-     "execution_count": 13,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_full_pandas"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We then **load the data into the getML engine**. We begin by setting a project."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-community-edition-linux...\n",
-      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240912153535.log.\n",
-      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
+    "cells": [
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "tags": [
+                    "hide_input"
+                ]
+            },
+            "source": [
+                "# Propositionalization: Predicting air pollution in Beijing\n",
+                "\n",
+                "In this notebook we will compare getML to featuretools and tsfresh, both of which open-source libraries for feature engineering. We find that advanced algorithms featured in getML yield significantly better predictions on this dataset. We then discuss why that is.\n",
+                "\n",
+                "Summary:\n",
+                "\n",
+                "- Prediction type: __Regression model__\n",
+                "- Domain: __Air pollution__\n",
+                "- Prediction target: __pm 2.5 concentration__ \n",
+                "- Source data: __Multivariate time series__\n",
+                "- Population size: __41757__"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "tags": [
+                    "remove_cell_on_docs"
+                ]
+            },
+            "source": [
+                "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/air_pollution_prop.ipynb\">\n",
+                "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+                "</a>"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Background\n",
+                "\n",
+                "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
+                "\n",
+                "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
+                "\n",
+                "As our example dataset, we use a publicly available dataset on air pollution in Beijing, China (https://archive.ics.uci.edu/dataset/381/beijing+pm2+5+data). For further details about the data set refer to [the full notebook](../air_pollution.ipynb)."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Analysis"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Table of contents\n",
+                "\n",
+                "1. [Loading data](#1.-Loading-data)\n",
+                "2. [Predictive modeling](#2.-Predictive-modeling)\n",
+                "3. [Comparison](#3.-Comparison)\n",
+                "4. [Conclusion](#4.-Conclusion)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 1,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Note: you may need to restart the kernel to use updated packages.\n"
+                    ]
+                }
+            ],
+            "source": [
+                "%pip install -q \"getml==1.5.0\" \"featuretools==1.31.0\" \"tsfresh==0.20.3\""
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 2,
+            "metadata": {
+                "tags": []
+            },
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "getML API version: 1.5.0\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "import os\n",
+                "import sys\n",
+                "\n",
+                "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
+                "from pathlib import Path\n",
+                "from urllib import request\n",
+                "import getml\n",
+                "import pandas as pd\n",
+                "\n",
+                "print(f\"getML API version: {getml.__version__}\\n\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 3,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+                "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+                "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 4,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "parent = Path(os.getcwd()).parent.as_posix()\n",
+                "\n",
+                "if parent not in sys.path:\n",
+                "    sys.path.append(parent)\n",
+                "\n",
+                "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 1. Loading data"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 1.1 Download from source\n",
+                "\n",
+                "We begin by downloading the data from the UCI Machine Learning repository."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 5,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "FEATURETOOLS_FILES = [\"featuretools_training.csv\", \"featuretools_test.csv\"]\n",
+                "\n",
+                "for fname in FEATURETOOLS_FILES:\n",
+                "    if not os.path.exists(fname):\n",
+                "        fname, res = request.urlretrieve(\n",
+                "            \"https://static.getml.com/datasets/air_pollution/featuretools/\" + fname,\n",
+                "            fname,\n",
+                "        )"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 6,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "TSFRESH_FILES = [\"tsfresh_training.csv\", \"tsfresh_test.csv\"]\n",
+                "\n",
+                "for fname in TSFRESH_FILES:\n",
+                "    if not os.path.exists(fname):\n",
+                "        fname, res = request.urlretrieve(\n",
+                "            \"https://static.getml.com/datasets/air_pollution/tsfresh/\" + fname, fname\n",
+                "        )"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 7,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "fname = \"PRSA_data_2010.1.1-2014.12.31.csv\"\n",
+                "\n",
+                "if not os.path.exists(fname):\n",
+                "    fname, res = request.urlretrieve(\n",
+                "        \"https://archive.ics.uci.edu/ml/machine-learning-databases/00381/\" + fname,\n",
+                "        fname,\n",
+                "    )"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 1.2 Prepare data for tsfresh and getML\n",
+                "\n",
+                "Our our goal is to predict the pm2.5 concentration from factors such as weather or time of day. However, there are some **missing entries** for pm2.5, so we get rid of them."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 8,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_full_pandas = pd.read_csv(fname)\n",
+                "\n",
+                "data_full_pandas = data_full_pandas[\n",
+                "    data_full_pandas[\"pm2.5\"] == data_full_pandas[\"pm2.5\"]\n",
+                "]"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "tsfresh requires a date column, so we build one."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 9,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "def add_leading_zero(val):\n",
+                "    if len(str(val)) == 1:\n",
+                "        return \"0\" + str(val)\n",
+                "    return str(val)\n",
+                "\n",
+                "\n",
+                "data_full_pandas[\"month\"] = [add_leading_zero(val) for val in data_full_pandas[\"month\"]]\n",
+                "\n",
+                "data_full_pandas[\"day\"] = [add_leading_zero(val) for val in data_full_pandas[\"day\"]]\n",
+                "\n",
+                "data_full_pandas[\"hour\"] = [add_leading_zero(val) for val in data_full_pandas[\"hour\"]]\n",
+                "\n",
+                "\n",
+                "def make_date(year, month, day, hour):\n",
+                "    return year + \"-\" + month + \"-\" + day + \" \" + hour + \":00:00\"\n",
+                "\n",
+                "\n",
+                "data_full_pandas[\"date\"] = [\n",
+                "    make_date(str(year), month, day, hour)\n",
+                "    for year, month, day, hour in zip(\n",
+                "        data_full_pandas[\"year\"],\n",
+                "        data_full_pandas[\"month\"],\n",
+                "        data_full_pandas[\"day\"],\n",
+                "        data_full_pandas[\"hour\"],\n",
+                "    )\n",
+                "]"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "tsfresh also requires the time series to have ids. Since there is only a single time series, that series has the same id."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 10,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_full_pandas[\"id\"] = 1"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The dataset now contains many columns that we do not need or that tsfresh cannot process. For instance, *cbwd* might actually contain useful information, but it is a categorical variable, which is difficult to handle for tsfresh, so we remove it."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We also want to split our data into a training and testing set."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 11,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_train_pandas = data_full_pandas[data_full_pandas[\"year\"] < 2014]\n",
+                "data_test_pandas = data_full_pandas[data_full_pandas[\"year\"] == 2014]\n",
+                "data_full_pandas = data_full_pandas"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 12,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "def remove_unwanted_columns(df):\n",
+                "    del df[\"cbwd\"]\n",
+                "    del df[\"year\"]\n",
+                "    del df[\"month\"]\n",
+                "    del df[\"day\"]\n",
+                "    del df[\"hour\"]\n",
+                "    del df[\"No\"]\n",
+                "    return df\n",
+                "\n",
+                "\n",
+                "data_full_pandas = remove_unwanted_columns(data_full_pandas)\n",
+                "data_train_pandas = remove_unwanted_columns(data_train_pandas)\n",
+                "data_test_pandas = remove_unwanted_columns(data_test_pandas)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 13,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>pm2.5</th>\n",
+                            "      <th>DEWP</th>\n",
+                            "      <th>TEMP</th>\n",
+                            "      <th>PRES</th>\n",
+                            "      <th>Iws</th>\n",
+                            "      <th>Is</th>\n",
+                            "      <th>Ir</th>\n",
+                            "      <th>date</th>\n",
+                            "      <th>id</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>24</th>\n",
+                            "      <td>129.0</td>\n",
+                            "      <td>-16</td>\n",
+                            "      <td>-4.0</td>\n",
+                            "      <td>1020.0</td>\n",
+                            "      <td>1.79</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 00:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>25</th>\n",
+                            "      <td>148.0</td>\n",
+                            "      <td>-15</td>\n",
+                            "      <td>-4.0</td>\n",
+                            "      <td>1020.0</td>\n",
+                            "      <td>2.68</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 01:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>26</th>\n",
+                            "      <td>159.0</td>\n",
+                            "      <td>-11</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1021.0</td>\n",
+                            "      <td>3.57</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 02:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>27</th>\n",
+                            "      <td>181.0</td>\n",
+                            "      <td>-7</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1022.0</td>\n",
+                            "      <td>5.36</td>\n",
+                            "      <td>1</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 03:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>28</th>\n",
+                            "      <td>138.0</td>\n",
+                            "      <td>-7</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1022.0</td>\n",
+                            "      <td>6.25</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 04:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>43819</th>\n",
+                            "      <td>8.0</td>\n",
+                            "      <td>-23</td>\n",
+                            "      <td>-2.0</td>\n",
+                            "      <td>1034.0</td>\n",
+                            "      <td>231.97</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2014-12-31 19:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>43820</th>\n",
+                            "      <td>10.0</td>\n",
+                            "      <td>-22</td>\n",
+                            "      <td>-3.0</td>\n",
+                            "      <td>1034.0</td>\n",
+                            "      <td>237.78</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2014-12-31 20:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>43821</th>\n",
+                            "      <td>10.0</td>\n",
+                            "      <td>-22</td>\n",
+                            "      <td>-3.0</td>\n",
+                            "      <td>1034.0</td>\n",
+                            "      <td>242.70</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2014-12-31 21:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>43822</th>\n",
+                            "      <td>8.0</td>\n",
+                            "      <td>-22</td>\n",
+                            "      <td>-4.0</td>\n",
+                            "      <td>1034.0</td>\n",
+                            "      <td>246.72</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2014-12-31 22:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>43823</th>\n",
+                            "      <td>12.0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>-3.0</td>\n",
+                            "      <td>1034.0</td>\n",
+                            "      <td>249.85</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2014-12-31 23:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>41757 rows × 9 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "       pm2.5  DEWP  TEMP    PRES     Iws  Is  Ir                 date  id\n",
+                            "24     129.0   -16  -4.0  1020.0    1.79   0   0  2010-01-02 00:00:00   1\n",
+                            "25     148.0   -15  -4.0  1020.0    2.68   0   0  2010-01-02 01:00:00   1\n",
+                            "26     159.0   -11  -5.0  1021.0    3.57   0   0  2010-01-02 02:00:00   1\n",
+                            "27     181.0    -7  -5.0  1022.0    5.36   1   0  2010-01-02 03:00:00   1\n",
+                            "28     138.0    -7  -5.0  1022.0    6.25   2   0  2010-01-02 04:00:00   1\n",
+                            "...      ...   ...   ...     ...     ...  ..  ..                  ...  ..\n",
+                            "43819    8.0   -23  -2.0  1034.0  231.97   0   0  2014-12-31 19:00:00   1\n",
+                            "43820   10.0   -22  -3.0  1034.0  237.78   0   0  2014-12-31 20:00:00   1\n",
+                            "43821   10.0   -22  -3.0  1034.0  242.70   0   0  2014-12-31 21:00:00   1\n",
+                            "43822    8.0   -22  -4.0  1034.0  246.72   0   0  2014-12-31 22:00:00   1\n",
+                            "43823   12.0   -21  -3.0  1034.0  249.85   0   0  2014-12-31 23:00:00   1\n",
+                            "\n",
+                            "[41757 rows x 9 columns]"
+                        ]
+                    },
+                    "execution_count": 13,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_full_pandas"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We then **load the data into the getML engine**. We begin by setting a project."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 14,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-linux...\n",
+                        "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240913121328.log.\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'air_pollution'</span>.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Connected to project \u001b[32m'air_pollution'\u001b[0m.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
+                "getml.engine.set_project(\"air_pollution\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 15,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_full = getml.data.DataFrame.from_pandas(data_full_pandas, name=\"full\")\n",
+                "df_full[\"date\"] = df_full[\"date\"].as_ts()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We need to **assign roles** to the columns, such as defining the target column."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 16,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th> name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"time_stamp\">                       date</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"target\"> pm2.5</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">     DEWP</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">     TEMP</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">     PRES</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">      Iws</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">       Is</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">       Ir</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"unused_float\">          id</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">                 time_stamp</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">target</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">unused_float</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> unit</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">time stamp, comparison only</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">      </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">            </th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2010-01-02</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">129</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-16</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1020</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.79</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2010-01-02 01:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">148</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-15</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1020</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">2</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.68</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2010-01-02 02:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">159</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-11</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-5</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1021</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">3</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.57</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2010-01-02 03:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">181</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-7</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-5</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1022</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">5</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.36</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2010-01-02 04:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">138</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-7</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-5</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1022</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">6</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.25</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>41752</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2014-12-31 19:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">8</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-23</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1034</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">231</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.97</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>41753</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2014-12-31 20:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">10</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-22</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-3</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1034</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">237</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.78</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>41754</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2014-12-31 21:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">10</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-22</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-3</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1034</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">242</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.7</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>41755</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2014-12-31 22:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">8</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-22</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1034</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">246</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.72</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>41756</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2014-12-31 23:00:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">12</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-21</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-3</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1034</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">249</span\n",
+                            "              ><span class=\"right\" style=\"width: 2ch\"\n",
+                            "                >.85</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    41757 rows x 9 columns<br />\n",
+                            "    memory usage: 3.01 MB<br />\n",
+                            "    name: full<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            " name                          date    pm2.5        DEWP        TEMP        PRES         Iws          Is          Ir\n",
+                            " role                    time_stamp   target   numerical   numerical   numerical   numerical   numerical   numerical\n",
+                            " unit   time stamp, comparison only                                                                                 \n",
+                            "    0                    2010-01-02      129        -16          -4         1020        1.79           0           0\n",
+                            "    1           2010-01-02 01:00:00      148        -15          -4         1020        2.68           0           0\n",
+                            "    2           2010-01-02 02:00:00      159        -11          -5         1021        3.57           0           0\n",
+                            "    3           2010-01-02 03:00:00      181         -7          -5         1022        5.36           1           0\n",
+                            "    4           2010-01-02 04:00:00      138         -7          -5         1022        6.25           2           0\n",
+                            "                                ...      ...         ...         ...         ...       ...           ...         ...\n",
+                            "41752           2014-12-31 19:00:00        8        -23          -2         1034      231.97           0           0\n",
+                            "41753           2014-12-31 20:00:00       10        -22          -3         1034      237.78           0           0\n",
+                            "41754           2014-12-31 21:00:00       10        -22          -3         1034      242.7            0           0\n",
+                            "41755           2014-12-31 22:00:00        8        -22          -4         1034      246.72           0           0\n",
+                            "41756           2014-12-31 23:00:00       12        -21          -3         1034      249.85           0           0\n",
+                            "\n",
+                            " name             id\n",
+                            " role   unused_float\n",
+                            " unit               \n",
+                            "    0              1\n",
+                            "    1              1\n",
+                            "    2              1\n",
+                            "    3              1\n",
+                            "    4              1\n",
+                            "                 ...\n",
+                            "41752              1\n",
+                            "41753              1\n",
+                            "41754              1\n",
+                            "41755              1\n",
+                            "41756              1\n",
+                            "\n",
+                            "\n",
+                            "41757 rows x 9 columns\n",
+                            "memory usage: 3.01 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 16,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "df_full.set_role([\"date\"], getml.data.roles.time_stamp)\n",
+                "df_full.set_role([\"pm2.5\"], getml.data.roles.target)\n",
+                "df_full.set_role(\n",
+                "    [\"DEWP\", \"TEMP\", \"PRES\", \"Iws\", \"Is\", \"Ir\"], getml.data.roles.numerical\n",
+                ")\n",
+                "df_full"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 17,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.numerical {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.numerical {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  td.char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th>     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th>     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    41757 rows<br />\n",
+                            "    \n",
+                            "    type: StringColumnView<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            "             \n",
+                            "    0   train\n",
+                            "    1   train\n",
+                            "    2   train\n",
+                            "    3   train\n",
+                            "    4   train\n",
+                            "        ...  \n",
+                            "\n",
+                            "\n",
+                            "41757 rows\n",
+                            "type: StringColumnView"
+                        ]
+                    },
+                    "execution_count": 17,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "split = getml.data.split.time(\n",
+                "    population=df_full, time_stamp=\"date\", test=getml.data.time.datetime(2014, 1, 1)\n",
+                ")\n",
+                "split"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 2. Predictive modeling"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.1 Propositionalization with getML's FastProp"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 18,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
+                            "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
+                            "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">full</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >date &lt;= date</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 1.0 days</tspan></text></svg></div>\n",
+                            "            </div>\n",
+                            "\n",
+                            "            <div style='margin-top: 15px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
+                            "            <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">data frames</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">staging table              </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">FULL__STAGING_TABLE_2</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "            </div>\n",
+                            "            \n",
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
+                            "<div style='margin-top: 15px;'>\n",
+                            "<div style='float: left; margin-right: 50px;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">subset</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">rows   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "<div style='float: left;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> rows</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">41757</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DataFrame</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "data model\n",
+                            "\n",
+                            "  population:\n",
+                            "    columns:\n",
+                            "    - DEWP: numerical\n",
+                            "    - TEMP: numerical\n",
+                            "    - PRES: numerical\n",
+                            "    - Iws: numerical\n",
+                            "    - Is: numerical\n",
+                            "    - ...\n",
+                            "\n",
+                            "    joins:\n",
+                            "    - right: 'full'\n",
+                            "      time_stamps: (population.date, full.date)\n",
+                            "      relationship: 'many-to-many'\n",
+                            "      memory: 86400.0\n",
+                            "      lagged_targets: False\n",
+                            "\n",
+                            "  full:\n",
+                            "    columns:\n",
+                            "    - DEWP: numerical\n",
+                            "    - TEMP: numerical\n",
+                            "    - PRES: numerical\n",
+                            "    - Iws: numerical\n",
+                            "    - Is: numerical\n",
+                            "    - ...\n",
+                            "\n",
+                            "\n",
+                            "container\n",
+                            "\n",
+                            "  population\n",
+                            "      subset   name   rows      type\n",
+                            "  0   test     full   unknown   View\n",
+                            "  1   train    full   unknown   View\n",
+                            "\n",
+                            "  peripheral\n",
+                            "      name    rows   type     \n",
+                            "  0   full   41757   DataFrame"
+                        ]
+                    },
+                    "execution_count": 18,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "time_series = getml.data.TimeSeries(\n",
+                "    population=df_full,\n",
+                "    alias=\"population\",\n",
+                "    split=split,\n",
+                "    time_stamps=\"date\",\n",
+                "    memory=getml.data.time.days(1),\n",
+                ")\n",
+                "\n",
+                "time_series"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 19,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=['full'],\n",
+                            "         predictors=[],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['memory: 1d', 'simple features'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=['full'],\n",
+                            "         predictors=[],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['memory: 1d', 'simple features'])"
+                        ]
+                    },
+                    "execution_count": 19,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "fast_prop = getml.feature_learning.FastProp(\n",
+                "    loss_function=getml.feature_learning.loss_functions.SquareLoss,\n",
+                "    num_threads=1,\n",
+                "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
+                ")\n",
+                "\n",
+                "pipe_fp_fl = getml.pipeline.Pipeline(\n",
+                "    tags=[\"memory: 1d\", \"simple features\"],\n",
+                "    data_model=time_series.data_model,\n",
+                "    feature_learners=[fast_prop],\n",
+                ")\n",
+                "\n",
+                "pipe_fp_fl"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 20,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K⠼ Checking... ━━━━━━━━━━━━━━━━━━━━                      50% • 00:01"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "pipe_fp_fl.check(time_series.train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 21,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "benchmark = Benchmark()"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 22,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:08\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:08.079324.\n",
+                        "\n",
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"fastprop\"):\n",
+                "    pipe_fp_fl.fit(time_series.train)\n",
+                "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 23,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 24,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "predictor = getml.predictors.XGBoostRegressor()\n",
+                "\n",
+                "pipe_fp_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 25,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:28\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:28.352024.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])"
+                        ]
+                    },
+                    "execution_count": 25,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.fit(fastprop_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 26,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used      </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">     mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 12:14:20</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">38.3028</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">55.2472</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6438</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 12:14:20</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">44.2502</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">63.4168</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.5462</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used         target        mae       rmse   rsquared\n",
+                            "0   2024-09-13 12:14:20   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
+                            "1   2024-09-13 12:14:20   fastprop_test    pm2.5     44.2502    63.4168     0.5462"
+                        ]
+                    },
+                    "execution_count": 26,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.score(fastprop_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.2 Using featuretools\n",
+                "\n",
+                "To make things a bit easier, we have written a high-level wrapper around featuretools which we placed in a separate module (`utils`)."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 27,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "ft_builder = FTTimeSeriesBuilder(\n",
+                "    num_features=200,\n",
+                "    horizon=pd.Timedelta(days=0),\n",
+                "    memory=pd.Timedelta(days=1),\n",
+                "    column_id=\"id\",\n",
+                "    time_stamp=\"date\",\n",
+                "    target=\"pm2.5\",\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 28,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "featuretools: Trying features...\n"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Selecting the best out of 298 features...\n",
+                        "Time taken: 0h:35m:42.348872\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"featuretools\"):\n",
+                "    featuretools_training = ft_builder.fit(data_train_pandas)\n",
+                "\n",
+                "featuretools_test = ft_builder.transform(data_test_pandas)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 29,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_featuretools_training = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_training, name=\"featuretools_training\"\n",
+                ")\n",
+                "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_test, name=\"featuretools_test\"\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 30,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "def set_roles_featuretools(df):\n",
+                "    df[\"date\"] = df[\"date\"].as_ts()\n",
+                "    df.set_role([\"pm2.5\"], getml.data.roles.target)\n",
+                "    df.set_role([\"date\"], getml.data.roles.time_stamp)\n",
+                "    df.set_role(df.roles.unused, getml.data.roles.numerical)\n",
+                "    df.set_role([\"id\"], getml.data.roles.unused_float)\n",
+                "    return df\n",
+                "\n",
+                "\n",
+                "df_featuretools_training = set_roles_featuretools(df_featuretools_training)\n",
+                "df_featuretools_test = set_roles_featuretools(df_featuretools_test)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 31,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['featuretools', 'memory: 1d'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['featuretools', 'memory: 1d'])"
+                        ]
+                    },
+                    "execution_count": 31,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "predictor = getml.predictors.XGBoostRegressor()\n",
+                "\n",
+                "pipe_ft_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"featuretools\", \"memory: 1d\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_ft_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 32,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K⠦ Checking...                                            0% • 00:00"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.check(df_featuretools_training)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 33,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:10\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:10.205240.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['featuretools', 'memory: 1d'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['featuretools', 'memory: 1d'])"
+                        ]
+                    },
+                    "execution_count": 33,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.fit(df_featuretools_training)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 34,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used             </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">     mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 12:58:20</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_training</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">38.277</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">54.8781</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6506</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 12:58:20</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">43.9151</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">62.5672</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.5594</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used                target        mae       rmse   rsquared\n",
+                            "0   2024-09-13 12:58:20   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
+                            "1   2024-09-13 12:58:20   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
+                        ]
+                    },
+                    "execution_count": 34,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.score(df_featuretools_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.3 Using tsfresh\n",
+                "\n",
+                "tsfresh is a rather low-level library. To make things a bit easier, we have written a high-level wrapper which we placed in a separate module (`utils`).\n",
+                "\n",
+                "To limit the memory consumption, we undertake the following steps:\n",
+                "\n",
+                "- We limit ourselves to a memory of 1 day from any point in time. This is necessary, because tsfresh duplicates records for every time stamp. That means that looking back 7 days instead of one day, the memory consumption would be  seven times as high.\n",
+                "- We extract only tsfresh's **MinimalFCParameters** and **IndexBasedFCParameters** (the latter is a superset of **TimeBasedFCParameters**).\n",
+                "\n",
+                "In order to make sure that tsfresh's features can be compared to getML's features, we also do the following:\n",
+                "\n",
+                "- We apply tsfresh's built-in feature selection algorithm.\n",
+                "- Of the remaining features, we only keep the 40 features most correlated with the target (in terms of the absolute value of the correlation).\n",
+                "- We add the original columns as additional features.\n"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 35,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>pm2.5</th>\n",
+                            "      <th>DEWP</th>\n",
+                            "      <th>TEMP</th>\n",
+                            "      <th>PRES</th>\n",
+                            "      <th>Iws</th>\n",
+                            "      <th>Is</th>\n",
+                            "      <th>Ir</th>\n",
+                            "      <th>date</th>\n",
+                            "      <th>id</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>24</th>\n",
+                            "      <td>129.0</td>\n",
+                            "      <td>-16</td>\n",
+                            "      <td>-4.0</td>\n",
+                            "      <td>1020.0</td>\n",
+                            "      <td>1.79</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 00:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>25</th>\n",
+                            "      <td>148.0</td>\n",
+                            "      <td>-15</td>\n",
+                            "      <td>-4.0</td>\n",
+                            "      <td>1020.0</td>\n",
+                            "      <td>2.68</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 01:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>26</th>\n",
+                            "      <td>159.0</td>\n",
+                            "      <td>-11</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1021.0</td>\n",
+                            "      <td>3.57</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 02:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>27</th>\n",
+                            "      <td>181.0</td>\n",
+                            "      <td>-7</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1022.0</td>\n",
+                            "      <td>5.36</td>\n",
+                            "      <td>1</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 03:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>28</th>\n",
+                            "      <td>138.0</td>\n",
+                            "      <td>-7</td>\n",
+                            "      <td>-5.0</td>\n",
+                            "      <td>1022.0</td>\n",
+                            "      <td>6.25</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2010-01-02 04:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>35059</th>\n",
+                            "      <td>22.0</td>\n",
+                            "      <td>-19</td>\n",
+                            "      <td>7.0</td>\n",
+                            "      <td>1013.0</td>\n",
+                            "      <td>114.87</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2013-12-31 19:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>35060</th>\n",
+                            "      <td>18.0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>7.0</td>\n",
+                            "      <td>1014.0</td>\n",
+                            "      <td>119.79</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2013-12-31 20:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>35061</th>\n",
+                            "      <td>23.0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>7.0</td>\n",
+                            "      <td>1014.0</td>\n",
+                            "      <td>125.60</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2013-12-31 21:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>35062</th>\n",
+                            "      <td>20.0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>6.0</td>\n",
+                            "      <td>1014.0</td>\n",
+                            "      <td>130.52</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2013-12-31 22:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>35063</th>\n",
+                            "      <td>23.0</td>\n",
+                            "      <td>-20</td>\n",
+                            "      <td>7.0</td>\n",
+                            "      <td>1014.0</td>\n",
+                            "      <td>137.67</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2013-12-31 23:00:00</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>33096 rows × 9 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "       pm2.5  DEWP  TEMP    PRES     Iws  Is  Ir                 date  id\n",
+                            "24     129.0   -16  -4.0  1020.0    1.79   0   0  2010-01-02 00:00:00   1\n",
+                            "25     148.0   -15  -4.0  1020.0    2.68   0   0  2010-01-02 01:00:00   1\n",
+                            "26     159.0   -11  -5.0  1021.0    3.57   0   0  2010-01-02 02:00:00   1\n",
+                            "27     181.0    -7  -5.0  1022.0    5.36   1   0  2010-01-02 03:00:00   1\n",
+                            "28     138.0    -7  -5.0  1022.0    6.25   2   0  2010-01-02 04:00:00   1\n",
+                            "...      ...   ...   ...     ...     ...  ..  ..                  ...  ..\n",
+                            "35059   22.0   -19   7.0  1013.0  114.87   0   0  2013-12-31 19:00:00   1\n",
+                            "35060   18.0   -21   7.0  1014.0  119.79   0   0  2013-12-31 20:00:00   1\n",
+                            "35061   23.0   -21   7.0  1014.0  125.60   0   0  2013-12-31 21:00:00   1\n",
+                            "35062   20.0   -21   6.0  1014.0  130.52   0   0  2013-12-31 22:00:00   1\n",
+                            "35063   23.0   -20   7.0  1014.0  137.67   0   0  2013-12-31 23:00:00   1\n",
+                            "\n",
+                            "[33096 rows x 9 columns]"
+                        ]
+                    },
+                    "execution_count": 35,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_train_pandas"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 36,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:25<00:00,  1.55it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.67it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:23<00:00,  1.67it/s]\n"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Selecting the best out of 78 features...\n",
+                        "Time taken: 0h:1m:21.479083\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:03<00:00, 13.11it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.33it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.12it/s]\n"
+                    ]
+                }
+            ],
+            "source": [
+                "tsfresh_builder = TSFreshBuilder(\n",
+                "    num_features=200, memory=24, column_id=\"id\", time_stamp=\"date\", target=\"pm2.5\"\n",
+                ")\n",
+                "\n",
+                "with benchmark(\"tsfresh\"):\n",
+                "    tsfresh_training = tsfresh_builder.fit(data_train_pandas)\n",
+                "\n",
+                "tsfresh_test = tsfresh_builder.transform(data_test_pandas)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "tsfresh does not contain built-in machine learning algorithms. In order to ensure a fair comparison, we use the exact same machine learning algorithm we have also used for getML: An XGBoost regressor with all hyperparameters set to their default value.\n",
+                "\n",
+                "In order to do so, we load the tsfresh features into the getML engine."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 37,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_tsfresh_training = getml.data.DataFrame.from_pandas(\n",
+                "    tsfresh_training, name=\"tsfresh_training\"\n",
+                ")\n",
+                "df_tsfresh_test = getml.data.DataFrame.from_pandas(tsfresh_test, name=\"tsfresh_test\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "As usual, we need to set roles:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 38,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "def set_roles_tsfresh(df):\n",
+                "    df[\"date\"] = df[\"date\"].as_ts()\n",
+                "    df.set_role([\"pm2.5\"], getml.data.roles.target)\n",
+                "    df.set_role([\"date\"], getml.data.roles.time_stamp)\n",
+                "    df.set_role(df.roles.unused, getml.data.roles.numerical)\n",
+                "    df.set_role([\"id\"], getml.data.roles.unused_float)\n",
+                "    return df\n",
+                "\n",
+                "\n",
+                "df_tsfresh_training = set_roles_tsfresh(df_tsfresh_training)\n",
+                "df_tsfresh_test = set_roles_tsfresh(df_tsfresh_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "In this case, our pipeline is very simple. It only consists of a single XGBoostRegressor."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 39,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['tsfresh', 'memory: 1d'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['tsfresh', 'memory: 1d'])"
+                        ]
+                    },
+                    "execution_count": 39,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "predictor = getml.predictors.XGBoostRegressor()\n",
+                "\n",
+                "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"tsfresh\", \"memory: 1d\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_tsf_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 40,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.check(df_tsfresh_training)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 41,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:05\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:05.645952.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['tsfresh', 'memory: 1d'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['tsfresh', 'memory: 1d'])"
+                        ]
+                    },
+                    "execution_count": 41,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.fit(df_tsfresh_training)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 42,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used        </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">     mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:00:08</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_training</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">40.8917</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">57.9517</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6099</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:00:08</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">pm2.5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">47.1106</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">66.6</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.5015</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used           target        mae       rmse   rsquared\n",
+                            "0   2024-09-13 13:00:08   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
+                            "1   2024-09-13 13:00:08   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
+                        ]
+                    },
+                    "execution_count": 42,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.score(df_tsfresh_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 3. Comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 43,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>runtime</th>\n",
+                            "      <th>num_features</th>\n",
+                            "      <th>features_per_second</th>\n",
+                            "      <th>normalized_runtime</th>\n",
+                            "      <th>normalized_runtime_per_feature</th>\n",
+                            "      <th>mae</th>\n",
+                            "      <th>rmse</th>\n",
+                            "      <th>rsquared</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>getML: FastProp</th>\n",
+                            "      <td>0 days 00:00:13.433118</td>\n",
+                            "      <td>289</td>\n",
+                            "      <td>21.514167</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>44.250202</td>\n",
+                            "      <td>63.416780</td>\n",
+                            "      <td>0.546191</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>featuretools</th>\n",
+                            "      <td>0 days 00:35:42.350566</td>\n",
+                            "      <td>114</td>\n",
+                            "      <td>0.053213</td>\n",
+                            "      <td>159.482747</td>\n",
+                            "      <td>404.306039</td>\n",
+                            "      <td>43.915071</td>\n",
+                            "      <td>62.567175</td>\n",
+                            "      <td>0.559369</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>tsfresh</th>\n",
+                            "      <td>0 days 00:01:21.479423</td>\n",
+                            "      <td>72</td>\n",
+                            "      <td>0.883658</td>\n",
+                            "      <td>6.065563</td>\n",
+                            "      <td>24.346701</td>\n",
+                            "      <td>47.110594</td>\n",
+                            "      <td>66.599982</td>\n",
+                            "      <td>0.501524</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "                               runtime  num_features  features_per_second  \\\n",
+                            "getML: FastProp 0 days 00:00:13.433118           289            21.514167   \n",
+                            "featuretools    0 days 00:35:42.350566           114             0.053213   \n",
+                            "tsfresh         0 days 00:01:21.479423            72             0.883658   \n",
+                            "\n",
+                            "                 normalized_runtime  normalized_runtime_per_feature  \\\n",
+                            "getML: FastProp            1.000000                        1.000000   \n",
+                            "featuretools             159.482747                      404.306039   \n",
+                            "tsfresh                    6.065563                       24.346701   \n",
+                            "\n",
+                            "                       mae       rmse  rsquared  \n",
+                            "getML: FastProp  44.250202  63.416780  0.546191  \n",
+                            "featuretools     43.915071  62.567175  0.559369  \n",
+                            "tsfresh          47.110594  66.599982  0.501524  "
+                        ]
+                    },
+                    "execution_count": 43,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "num_features = dict(\n",
+                "    fastprop=289,\n",
+                "    featuretools=114,\n",
+                "    tsfresh=72,\n",
+                ")\n",
+                "\n",
+                "runtime_per_feature = [\n",
+                "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
+                "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
+                "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
+                "]\n",
+                "\n",
+                "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
+                "\n",
+                "normalized_runtime_per_feature = [\n",
+                "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
+                "]\n",
+                "\n",
+                "comparison = pd.DataFrame(\n",
+                "    dict(\n",
+                "        runtime=[\n",
+                "            benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"featuretools\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"],\n",
+                "        ],\n",
+                "        num_features=num_features.values(),\n",
+                "        features_per_second=features_per_second,\n",
+                "        normalized_runtime=[\n",
+                "            1,\n",
+                "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "        ],\n",
+                "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
+                "        mae=[pipe_fp_pr.mae, pipe_ft_pr.mae, pipe_tsf_pr.mae],\n",
+                "        rmse=[pipe_fp_pr.rmse, pipe_ft_pr.rmse, pipe_tsf_pr.rmse],\n",
+                "        rsquared=[pipe_fp_pr.rsquared, pipe_ft_pr.rsquared, pipe_tsf_pr.rsquared],\n",
+                "    )\n",
+                ")\n",
+                "\n",
+                "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]\n",
+                "\n",
+                "comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 44,
+            "metadata": {
+                "tags": []
+            },
+            "outputs": [],
+            "source": [
+                "# export for further use\n",
+                "comparison.to_csv(\"comparisons/air_pollution.csv\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 4. Conclusion\n",
+                "\n",
+                "We have compared getML's feature learning algorithms to tsfresh's brute-force feature engineering approaches on a data set related to air pollution in China. We found that getML significantly outperforms featuretools and tsfresh. These results are consistent with the view that feature learning can yield significant improvements over simple propositionalization approaches.\n",
+                "\n",
+                "However, there are other datasets on which simple propositionalization performs well. Our suggestion is therefore to think of algorithms like `FastProp` and `RelMT` as tools in a toolbox. If a simple tool like `FastProp` gets the job done, then use that. But when you need more advanced approaches, like `RelMT`, you should have them at your disposal as well.\n",
+                "\n",
+                "You are encouraged to reproduce these results."
+            ]
+        }
+    ],
+    "metadata": {
+        "celltoolbar": "Tags",
+        "kernelspec": {
+            "display_name": "Python 3 (ipykernel)",
+            "language": "python",
+            "name": "python3"
+        },
+        "language_info": {
+            "codemirror_mode": {
+                "name": "ipython",
+                "version": 3
+            },
+            "file_extension": ".py",
+            "mimetype": "text/x-python",
+            "name": "python",
+            "nbconvert_exporter": "python",
+            "pygments_lexer": "ipython3",
+            "version": "3.11.9"
+        }
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'air_pollution'</span>.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'air_pollution'\u001b[0m.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
-    "getml.engine.set_project(\"air_pollution\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_full = getml.data.DataFrame.from_pandas(data_full_pandas, name=\"full\")\n",
-    "df_full[\"date\"] = df_full[\"date\"].as_ts()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We need to **assign roles** to the columns, such as defining the target column."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th> name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"time_stamp\">                       date</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"target\"> pm2.5</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">     DEWP</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">     TEMP</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">     PRES</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">      Iws</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">       Is</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">       Ir</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"unused_float\">          id</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">                 time_stamp</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">target</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">unused_float</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> unit</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">time stamp, comparison only</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">      </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">            </th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2010-01-02</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">129</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-16</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1020</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.79</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2010-01-02 01:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">148</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-15</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1020</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">2</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.68</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2010-01-02 02:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">159</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-11</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-5</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1021</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">3</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.57</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2010-01-02 03:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">181</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-7</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-5</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1022</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">5</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.36</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2010-01-02 04:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">138</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-7</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-5</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1022</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">6</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.25</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">...</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>41752</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2014-12-31 19:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">8</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-23</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1034</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">231</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.97</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>41753</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2014-12-31 20:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">10</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-22</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-3</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1034</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">237</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.78</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>41754</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2014-12-31 21:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">10</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-22</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-3</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1034</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">242</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.7</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>41755</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2014-12-31 22:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">8</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-22</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1034</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">246</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.72</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>41756</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2014-12-31 23:00:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">12</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-21</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-3</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1034</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">249</span\n",
-       "              ><span class=\"right\" style=\"width: 2ch\"\n",
-       "                >.85</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    41757 rows x 9 columns<br />\n",
-       "    memory usage: 3.01 MB<br />\n",
-       "    name: full<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       " name                          date    pm2.5        DEWP        TEMP        PRES         Iws          Is          Ir\n",
-       " role                    time_stamp   target   numerical   numerical   numerical   numerical   numerical   numerical\n",
-       " unit   time stamp, comparison only                                                                                 \n",
-       "    0                    2010-01-02      129        -16          -4         1020        1.79           0           0\n",
-       "    1           2010-01-02 01:00:00      148        -15          -4         1020        2.68           0           0\n",
-       "    2           2010-01-02 02:00:00      159        -11          -5         1021        3.57           0           0\n",
-       "    3           2010-01-02 03:00:00      181         -7          -5         1022        5.36           1           0\n",
-       "    4           2010-01-02 04:00:00      138         -7          -5         1022        6.25           2           0\n",
-       "                                ...      ...         ...         ...         ...       ...           ...         ...\n",
-       "41752           2014-12-31 19:00:00        8        -23          -2         1034      231.97           0           0\n",
-       "41753           2014-12-31 20:00:00       10        -22          -3         1034      237.78           0           0\n",
-       "41754           2014-12-31 21:00:00       10        -22          -3         1034      242.7            0           0\n",
-       "41755           2014-12-31 22:00:00        8        -22          -4         1034      246.72           0           0\n",
-       "41756           2014-12-31 23:00:00       12        -21          -3         1034      249.85           0           0\n",
-       "\n",
-       " name             id\n",
-       " role   unused_float\n",
-       " unit               \n",
-       "    0              1\n",
-       "    1              1\n",
-       "    2              1\n",
-       "    3              1\n",
-       "    4              1\n",
-       "                 ...\n",
-       "41752              1\n",
-       "41753              1\n",
-       "41754              1\n",
-       "41755              1\n",
-       "41756              1\n",
-       "\n",
-       "\n",
-       "41757 rows x 9 columns\n",
-       "memory usage: 3.01 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 16,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df_full.set_role([\"date\"], getml.data.roles.time_stamp)\n",
-    "df_full.set_role([\"pm2.5\"], getml.data.roles.target)\n",
-    "df_full.set_role(\n",
-    "    [\"DEWP\", \"TEMP\", \"PRES\", \"Iws\", \"Is\", \"Ir\"], getml.data.roles.numerical\n",
-    ")\n",
-    "df_full"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.numerical {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.numerical {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  td.char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th>     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th>     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td>...</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    41757 rows<br />\n",
-       "    \n",
-       "    type: StringColumnView<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       "             \n",
-       "    0   train\n",
-       "    1   train\n",
-       "    2   train\n",
-       "    3   train\n",
-       "    4   train\n",
-       "        ...  \n",
-       "\n",
-       "\n",
-       "41757 rows\n",
-       "type: StringColumnView"
-      ]
-     },
-     "execution_count": 17,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "split = getml.data.split.time(\n",
-    "    population=df_full, time_stamp=\"date\", test=getml.data.time.datetime(2014, 1, 1)\n",
-    ")\n",
-    "split"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 2. Predictive modeling"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.1 Propositionalization with getML's FastProp"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
-       "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
-       "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">full</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >date &lt;= date</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 1.0 days</tspan></text></svg></div>\n",
-       "            </div>\n",
-       "\n",
-       "            <div style='margin-top: 15px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
-       "            <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">data frames</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">staging table              </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">FULL__STAGING_TABLE_2</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "            </div>\n",
-       "            \n",
-       "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
-       "<div style='margin-top: 15px;'>\n",
-       "<div style='float: left; margin-right: 50px;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">subset</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">rows   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "<div style='float: left;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> rows</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">41757</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DataFrame</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "data model\n",
-       "\n",
-       "  population:\n",
-       "    columns:\n",
-       "    - DEWP: numerical\n",
-       "    - TEMP: numerical\n",
-       "    - PRES: numerical\n",
-       "    - Iws: numerical\n",
-       "    - Is: numerical\n",
-       "    - ...\n",
-       "\n",
-       "    joins:\n",
-       "    - right: 'full'\n",
-       "      time_stamps: (population.date, full.date)\n",
-       "      relationship: 'many-to-many'\n",
-       "      memory: 86400.0\n",
-       "      lagged_targets: False\n",
-       "\n",
-       "  full:\n",
-       "    columns:\n",
-       "    - DEWP: numerical\n",
-       "    - TEMP: numerical\n",
-       "    - PRES: numerical\n",
-       "    - Iws: numerical\n",
-       "    - Is: numerical\n",
-       "    - ...\n",
-       "\n",
-       "\n",
-       "container\n",
-       "\n",
-       "  population\n",
-       "      subset   name   rows      type\n",
-       "  0   test     full   unknown   View\n",
-       "  1   train    full   unknown   View\n",
-       "\n",
-       "  peripheral\n",
-       "      name    rows   type     \n",
-       "  0   full   41757   DataFrame"
-      ]
-     },
-     "execution_count": 18,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "time_series = getml.data.TimeSeries(\n",
-    "    population=df_full,\n",
-    "    alias=\"population\",\n",
-    "    split=split,\n",
-    "    time_stamps=\"date\",\n",
-    "    memory=getml.data.time.days(1),\n",
-    ")\n",
-    "\n",
-    "time_series"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=['full'],\n",
-       "         predictors=[],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['memory: 1d', 'simple features'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=['full'],\n",
-       "         predictors=[],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['memory: 1d', 'simple features'])"
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "fast_prop = getml.feature_learning.FastProp(\n",
-    "    loss_function=getml.feature_learning.loss_functions.SquareLoss,\n",
-    "    num_threads=1,\n",
-    "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
-    ")\n",
-    "\n",
-    "pipe_fp_fl = getml.pipeline.Pipeline(\n",
-    "    tags=[\"memory: 1d\", \"simple features\"],\n",
-    "    data_model=time_series.data_model,\n",
-    "    feature_learners=[fast_prop],\n",
-    ")\n",
-    "\n",
-    "pipe_fp_fl"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pipe_fp_fl.check(time_series.train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "benchmark = Benchmark()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:04.439985.\n",
-      "\n",
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"fastprop\"):\n",
-    "    pipe_fp_fl.fit(time_series.train)\n",
-    "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "predictor = getml.predictors.XGBoostRegressor()\n",
-    "\n",
-    "pipe_fp_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:15\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:15.893621.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])"
-      ]
-     },
-     "execution_count": 25,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.fit(fastprop_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used      </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">     mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 15:36:02</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">38.3028</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">55.2472</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6438</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 15:36:02</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">44.2526</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">63.4191</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.5462</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used         target        mae       rmse   rsquared\n",
-       "0   2024-09-12 15:36:02   fastprop_train   pm2.5     38.3028    55.2472     0.6438\n",
-       "1   2024-09-12 15:36:02   fastprop_test    pm2.5     44.2526    63.4191     0.5462"
-      ]
-     },
-     "execution_count": 26,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.score(fastprop_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.2 Using featuretools\n",
-    "\n",
-    "To make things a bit easier, we have written a high-level wrapper around featuretools which we placed in a separate module (`utils`)."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "ft_builder = FTTimeSeriesBuilder(\n",
-    "    num_features=200,\n",
-    "    horizon=pd.Timedelta(days=0),\n",
-    "    memory=pd.Timedelta(days=1),\n",
-    "    column_id=\"id\",\n",
-    "    time_stamp=\"date\",\n",
-    "    target=\"pm2.5\",\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 28,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "featuretools: Trying features...\n",
-      "Selecting the best out of 298 features...\n",
-      "Time taken: 0h:29m:33.992379\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"featuretools\"):\n",
-    "    featuretools_training = ft_builder.fit(data_train_pandas)\n",
-    "\n",
-    "featuretools_test = ft_builder.transform(data_test_pandas)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 29,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_featuretools_training = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_training, name=\"featuretools_training\"\n",
-    ")\n",
-    "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_test, name=\"featuretools_test\"\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def set_roles_featuretools(df):\n",
-    "    df[\"date\"] = df[\"date\"].as_ts()\n",
-    "    df.set_role([\"pm2.5\"], getml.data.roles.target)\n",
-    "    df.set_role([\"date\"], getml.data.roles.time_stamp)\n",
-    "    df.set_role(df.roles.unused, getml.data.roles.numerical)\n",
-    "    df.set_role([\"id\"], getml.data.roles.unused_float)\n",
-    "    return df\n",
-    "\n",
-    "\n",
-    "df_featuretools_training = set_roles_featuretools(df_featuretools_training)\n",
-    "df_featuretools_test = set_roles_featuretools(df_featuretools_test)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 31,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['featuretools', 'memory: 1d'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['featuretools', 'memory: 1d'])"
-      ]
-     },
-     "execution_count": 31,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "predictor = getml.predictors.XGBoostRegressor()\n",
-    "\n",
-    "pipe_ft_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"featuretools\", \"memory: 1d\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_ft_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.check(df_featuretools_training)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:10\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:10.147840.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['featuretools', 'memory: 1d'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['featuretools', 'memory: 1d'])"
-      ]
-     },
-     "execution_count": 33,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.fit(df_featuretools_training)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 34,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used             </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">     mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:13:39</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_training</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">38.277</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">54.8781</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6506</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:13:39</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">43.9151</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">62.5672</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.5594</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used                target        mae       rmse   rsquared\n",
-       "0   2024-09-12 16:13:39   featuretools_training   pm2.5     38.277     54.8781     0.6506\n",
-       "1   2024-09-12 16:13:39   featuretools_test       pm2.5     43.9151    62.5672     0.5594"
-      ]
-     },
-     "execution_count": 34,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.score(df_featuretools_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.3 Using tsfresh\n",
-    "\n",
-    "tsfresh is a rather low-level library. To make things a bit easier, we have written a high-level wrapper which we placed in a separate module (`utils`).\n",
-    "\n",
-    "To limit the memory consumption, we undertake the following steps:\n",
-    "\n",
-    "- We limit ourselves to a memory of 1 day from any point in time. This is necessary, because tsfresh duplicates records for every time stamp. That means that looking back 7 days instead of one day, the memory consumption would be  seven times as high.\n",
-    "- We extract only tsfresh's **MinimalFCParameters** and **IndexBasedFCParameters** (the latter is a superset of **TimeBasedFCParameters**).\n",
-    "\n",
-    "In order to make sure that tsfresh's features can be compared to getML's features, we also do the following:\n",
-    "\n",
-    "- We apply tsfresh's built-in feature selection algorithm.\n",
-    "- Of the remaining features, we only keep the 40 features most correlated with the target (in terms of the absolute value of the correlation).\n",
-    "- We add the original columns as additional features.\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 35,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>pm2.5</th>\n",
-       "      <th>DEWP</th>\n",
-       "      <th>TEMP</th>\n",
-       "      <th>PRES</th>\n",
-       "      <th>Iws</th>\n",
-       "      <th>Is</th>\n",
-       "      <th>Ir</th>\n",
-       "      <th>date</th>\n",
-       "      <th>id</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>24</th>\n",
-       "      <td>129.0</td>\n",
-       "      <td>-16</td>\n",
-       "      <td>-4.0</td>\n",
-       "      <td>1020.0</td>\n",
-       "      <td>1.79</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 00:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25</th>\n",
-       "      <td>148.0</td>\n",
-       "      <td>-15</td>\n",
-       "      <td>-4.0</td>\n",
-       "      <td>1020.0</td>\n",
-       "      <td>2.68</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 01:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>26</th>\n",
-       "      <td>159.0</td>\n",
-       "      <td>-11</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1021.0</td>\n",
-       "      <td>3.57</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 02:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>27</th>\n",
-       "      <td>181.0</td>\n",
-       "      <td>-7</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1022.0</td>\n",
-       "      <td>5.36</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 03:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>28</th>\n",
-       "      <td>138.0</td>\n",
-       "      <td>-7</td>\n",
-       "      <td>-5.0</td>\n",
-       "      <td>1022.0</td>\n",
-       "      <td>6.25</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2010-01-02 04:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35059</th>\n",
-       "      <td>22.0</td>\n",
-       "      <td>-19</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1013.0</td>\n",
-       "      <td>114.87</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2013-12-31 19:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35060</th>\n",
-       "      <td>18.0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1014.0</td>\n",
-       "      <td>119.79</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2013-12-31 20:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35061</th>\n",
-       "      <td>23.0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1014.0</td>\n",
-       "      <td>125.60</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2013-12-31 21:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35062</th>\n",
-       "      <td>20.0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>6.0</td>\n",
-       "      <td>1014.0</td>\n",
-       "      <td>130.52</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2013-12-31 22:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>35063</th>\n",
-       "      <td>23.0</td>\n",
-       "      <td>-20</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1014.0</td>\n",
-       "      <td>137.67</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2013-12-31 23:00:00</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>33096 rows × 9 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       pm2.5  DEWP  TEMP    PRES     Iws  Is  Ir                 date  id\n",
-       "24     129.0   -16  -4.0  1020.0    1.79   0   0  2010-01-02 00:00:00   1\n",
-       "25     148.0   -15  -4.0  1020.0    2.68   0   0  2010-01-02 01:00:00   1\n",
-       "26     159.0   -11  -5.0  1021.0    3.57   0   0  2010-01-02 02:00:00   1\n",
-       "27     181.0    -7  -5.0  1022.0    5.36   1   0  2010-01-02 03:00:00   1\n",
-       "28     138.0    -7  -5.0  1022.0    6.25   2   0  2010-01-02 04:00:00   1\n",
-       "...      ...   ...   ...     ...     ...  ..  ..                  ...  ..\n",
-       "35059   22.0   -19   7.0  1013.0  114.87   0   0  2013-12-31 19:00:00   1\n",
-       "35060   18.0   -21   7.0  1014.0  119.79   0   0  2013-12-31 20:00:00   1\n",
-       "35061   23.0   -21   7.0  1014.0  125.60   0   0  2013-12-31 21:00:00   1\n",
-       "35062   20.0   -21   6.0  1014.0  130.52   0   0  2013-12-31 22:00:00   1\n",
-       "35063   23.0   -20   7.0  1014.0  137.67   0   0  2013-12-31 23:00:00   1\n",
-       "\n",
-       "[33096 rows x 9 columns]"
-      ]
-     },
-     "execution_count": 35,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_train_pandas"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 36,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:22<00:00,  1.77it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:22<00:00,  1.79it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:22<00:00,  1.75it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Selecting the best out of 78 features...\n",
-      "Time taken: 0h:1m:15.803921\n",
-      "\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 16.82it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.35it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.05it/s]\n"
-     ]
-    }
-   ],
-   "source": [
-    "tsfresh_builder = TSFreshBuilder(\n",
-    "    num_features=200, memory=24, column_id=\"id\", time_stamp=\"date\", target=\"pm2.5\"\n",
-    ")\n",
-    "\n",
-    "with benchmark(\"tsfresh\"):\n",
-    "    tsfresh_training = tsfresh_builder.fit(data_train_pandas)\n",
-    "\n",
-    "tsfresh_test = tsfresh_builder.transform(data_test_pandas)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "tsfresh does not contain built-in machine learning algorithms. In order to ensure a fair comparison, we use the exact same machine learning algorithm we have also used for getML: An XGBoost regressor with all hyperparameters set to their default value.\n",
-    "\n",
-    "In order to do so, we load the tsfresh features into the getML engine."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 37,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_tsfresh_training = getml.data.DataFrame.from_pandas(\n",
-    "    tsfresh_training, name=\"tsfresh_training\"\n",
-    ")\n",
-    "df_tsfresh_test = getml.data.DataFrame.from_pandas(tsfresh_test, name=\"tsfresh_test\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As usual, we need to set roles:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 38,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def set_roles_tsfresh(df):\n",
-    "    df[\"date\"] = df[\"date\"].as_ts()\n",
-    "    df.set_role([\"pm2.5\"], getml.data.roles.target)\n",
-    "    df.set_role([\"date\"], getml.data.roles.time_stamp)\n",
-    "    df.set_role(df.roles.unused, getml.data.roles.numerical)\n",
-    "    df.set_role([\"id\"], getml.data.roles.unused_float)\n",
-    "    return df\n",
-    "\n",
-    "\n",
-    "df_tsfresh_training = set_roles_tsfresh(df_tsfresh_training)\n",
-    "df_tsfresh_test = set_roles_tsfresh(df_tsfresh_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "In this case, our pipeline is very simple. It only consists of a single XGBoostRegressor."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 39,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['tsfresh', 'memory: 1d'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['tsfresh', 'memory: 1d'])"
-      ]
-     },
-     "execution_count": 39,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "predictor = getml.predictors.XGBoostRegressor()\n",
-    "\n",
-    "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"tsfresh\", \"memory: 1d\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_tsf_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 40,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.check(df_tsfresh_training)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 41,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:05\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:05.349890.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['tsfresh', 'memory: 1d'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['tsfresh', 'memory: 1d'])"
-      ]
-     },
-     "execution_count": 41,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.fit(df_tsfresh_training)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 42,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used        </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">     mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:15:17</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_training</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">40.8917</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">57.9517</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6099</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:15:18</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">pm2.5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">47.1106</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">66.6</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.5015</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used           target        mae       rmse   rsquared\n",
-       "0   2024-09-12 16:15:17   tsfresh_training   pm2.5     40.8917    57.9517     0.6099\n",
-       "1   2024-09-12 16:15:18   tsfresh_test       pm2.5     47.1106    66.6        0.5015"
-      ]
-     },
-     "execution_count": 42,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.score(df_tsfresh_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 3. Comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 43,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>runtime</th>\n",
-       "      <th>num_features</th>\n",
-       "      <th>features_per_second</th>\n",
-       "      <th>normalized_runtime</th>\n",
-       "      <th>normalized_runtime_per_feature</th>\n",
-       "      <th>mae</th>\n",
-       "      <th>rmse</th>\n",
-       "      <th>rsquared</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:07.285902</td>\n",
-       "      <td>289</td>\n",
-       "      <td>39.665225</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>44.252635</td>\n",
-       "      <td>63.419113</td>\n",
-       "      <td>0.546164</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>featuretools</th>\n",
-       "      <td>0 days 00:29:33.994416</td>\n",
-       "      <td>114</td>\n",
-       "      <td>0.064262</td>\n",
-       "      <td>243.483156</td>\n",
-       "      <td>617.244655</td>\n",
-       "      <td>43.915071</td>\n",
-       "      <td>62.567175</td>\n",
-       "      <td>0.559369</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:01:15.804228</td>\n",
-       "      <td>72</td>\n",
-       "      <td>0.949816</td>\n",
-       "      <td>10.404234</td>\n",
-       "      <td>41.760977</td>\n",
-       "      <td>47.110594</td>\n",
-       "      <td>66.599982</td>\n",
-       "      <td>0.501524</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:07.285902           289            39.665225   \n",
-       "featuretools    0 days 00:29:33.994416           114             0.064262   \n",
-       "tsfresh         0 days 00:01:15.804228            72             0.949816   \n",
-       "\n",
-       "                 normalized_runtime  normalized_runtime_per_feature  \\\n",
-       "getML: FastProp            1.000000                        1.000000   \n",
-       "featuretools             243.483156                      617.244655   \n",
-       "tsfresh                   10.404234                       41.760977   \n",
-       "\n",
-       "                       mae       rmse  rsquared  \n",
-       "getML: FastProp  44.252635  63.419113  0.546164  \n",
-       "featuretools     43.915071  62.567175  0.559369  \n",
-       "tsfresh          47.110594  66.599982  0.501524  "
-      ]
-     },
-     "execution_count": 43,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "num_features = dict(\n",
-    "    fastprop=289,\n",
-    "    featuretools=114,\n",
-    "    tsfresh=72,\n",
-    ")\n",
-    "\n",
-    "runtime_per_feature = [\n",
-    "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
-    "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
-    "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
-    "]\n",
-    "\n",
-    "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
-    "\n",
-    "normalized_runtime_per_feature = [\n",
-    "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
-    "]\n",
-    "\n",
-    "comparison = pd.DataFrame(\n",
-    "    dict(\n",
-    "        runtime=[\n",
-    "            benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"featuretools\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"],\n",
-    "        ],\n",
-    "        num_features=num_features.values(),\n",
-    "        features_per_second=features_per_second,\n",
-    "        normalized_runtime=[\n",
-    "            1,\n",
-    "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "        ],\n",
-    "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
-    "        mae=[pipe_fp_pr.mae, pipe_ft_pr.mae, pipe_tsf_pr.mae],\n",
-    "        rmse=[pipe_fp_pr.rmse, pipe_ft_pr.rmse, pipe_tsf_pr.rmse],\n",
-    "        rsquared=[pipe_fp_pr.rsquared, pipe_ft_pr.rsquared, pipe_tsf_pr.rsquared],\n",
-    "    )\n",
-    ")\n",
-    "\n",
-    "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]\n",
-    "\n",
-    "comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 44,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "# export for further use\n",
-    "comparison.to_csv(\"comparisons/air_pollution.csv\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 4. Conclusion\n",
-    "\n",
-    "We have compared getML's feature learning algorithms to tsfresh's brute-force feature engineering approaches on a data set related to air pollution in China. We found that getML significantly outperforms featuretools and tsfresh. These results are consistent with the view that feature learning can yield significant improvements over simple propositionalization approaches.\n",
-    "\n",
-    "However, there are other datasets on which simple propositionalization performs well. Our suggestion is therefore to think of algorithms like `FastProp` and `RelMT` as tools in a toolbox. If a simple tool like `FastProp` gets the job done, then use that. But when you need more advanced approaches, like `RelMT`, you should have them at your disposal as well.\n",
-    "\n",
-    "You are encouraged to reproduce these results."
-   ]
-  }
- ],
- "metadata": {
-  "celltoolbar": "Tags",
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.9"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
+    "nbformat": 4,
+    "nbformat_minor": 4
 }
diff --git a/fastprop_benchmark/comparisons/air_pollution.csv b/fastprop_benchmark/comparisons/air_pollution.csv
index 62e1ecb..93690a9 100644
--- a/fastprop_benchmark/comparisons/air_pollution.csv
+++ b/fastprop_benchmark/comparisons/air_pollution.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,mae,rmse,rsquared
-getML: FastProp,0 days 00:00:07.285902,289,39.66522549680695,1.0,1.0,44.2526350869082,63.41911260454679,0.5461637964380341
-featuretools,0 days 00:29:33.994416,114,0.06426175612599289,243.4831563751475,617.2446551108643,43.91507134716814,62.56717509580391,0.5593690691075496
-tsfresh,0 days 00:01:15.804228,72,0.9498155458210015,10.40423382032863,41.76097735115624,47.11059400765295,66.59998183642728,0.5015240507138585
+getML: FastProp,0 days 00:00:13.433118,289,21.514167079021536,1.0,1.0,44.25020198393413,63.41678046742225,0.5461908346025002
+featuretools,0 days 00:35:42.350566,114,0.05321257909185177,159.4827474901955,404.30603902669907,43.91507134716814,62.56717509580391,0.5593690691075496
+tsfresh,0 days 00:01:21.479423,72,0.8836584165371371,6.065562961629608,24.34670080247843,47.11059400765295,66.59998183642728,0.5015240507138585
diff --git a/fastprop_benchmark/comparisons/dodgers.csv b/fastprop_benchmark/comparisons/dodgers.csv
index 6b7ad12..608489a 100644
--- a/fastprop_benchmark/comparisons/dodgers.csv
+++ b/fastprop_benchmark/comparisons/dodgers.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:12.036174,526,43.7024735600035,1.0,1.0,0.674739861410577,7.824273388224941,5.615137721839041
-featuretools,0 days 00:08:56.482944,59,0.10997554034007297,44.572548053891545,397.38357661043614,0.6497682308113026,8.500887091670293,6.08627656339764
-tsfresh,0 days 00:00:32.318518,12,0.3713041315010712,2.685115552500321,117.69993881653701,0.5778110797835911,8.913407825293008,6.788610043264059
+getML: FastProp,0 days 00:00:12.106112,526,43.449923962633065,1.0,1.0,0.674740264167421,7.824265418275141,5.615102199203448
+featuretools,0 days 00:08:51.202688,59,0.11106871988302243,43.87888431892915,391.1985661525092,0.6497682308113026,8.500887091670293,6.08627656339764
+tsfresh,0 days 00:00:31.919755,12,0.37594267626072375,2.636664438591019,115.5759287421247,0.5778110797835911,8.913407825293008,6.788610043264059
diff --git a/fastprop_benchmark/comparisons/interstate94.csv b/fastprop_benchmark/comparisons/interstate94.csv
index 118c1ba..9b6ebbf 100644
--- a/fastprop_benchmark/comparisons/interstate94.csv
+++ b/fastprop_benchmark/comparisons/interstate94.csv
@@ -1,3 +1,3 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:04.800454,461,96.03380389897244,1.0,1.0,0.9826778318692238,261.9388731680907,180.4867341518402
-featuretools,0 days 00:04:33.370925,59,0.21582395326461787,56.946889815005,444.9636031883223,0.9745821357660296,317.51997565190663,210.1987933667501
+getML: FastProp,0 days 00:00:04.806504,461,95.9140610013428,1.0,1.0,0.9826778318692238,261.9388731680907,180.4867341518402
+featuretools,0 days 00:04:27.009351,59,0.22096605475273678,55.551675604555825,434.06694801457894,0.9745821357660296,317.51997565190663,210.1987933667501
diff --git a/fastprop_benchmark/comparisons/occupancy.csv b/fastprop_benchmark/comparisons/occupancy.csv
index fc554fb..5b911dc 100644
--- a/fastprop_benchmark/comparisons/occupancy.csv
+++ b/fastprop_benchmark/comparisons/occupancy.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,accuracy,auc,cross_entropy
-getML: FastProp,0 days 00:00:01.854957,289,155.78750584203146,1.0,1.0,0.9888233849177106,0.9981659495576909,0.0442127968034932
-featuretools,0 days 00:07:28.129456,103,0.2298443195470412,241.58482164276583,677.7957625798412,0.988454925079833,0.9972073080154513,0.04923589235646241
-tsfresh,0 days 00:00:17.070465,60,3.51483965301503,9.202620330282588,44.32279171210469,0.9877180054040776,0.9978609436533588,0.04935860166671465
+getML: FastProp,0 days 00:00:01.825967,289,158.27793605571384,1.0,1.0,0.9888233849177106,0.9981659495576908,0.044212831751311917
+featuretools,0 days 00:07:20.110459,103,0.234032161167652,241.0287036950832,676.3084836973726,0.988454925079833,0.9972073080154513,0.04923589235646241
+tsfresh,0 days 00:00:14.295312,60,4.197183689744182,7.828899426988549,37.7105096549541,0.9877180054040776,0.9978609436533588,0.04935860166671465
diff --git a/fastprop_benchmark/comparisons/robot.csv b/fastprop_benchmark/comparisons/robot.csv
index 00aec93..bc7f2c6 100644
--- a/fastprop_benchmark/comparisons/robot.csv
+++ b/fastprop_benchmark/comparisons/robot.csv
@@ -1,4 +1,4 @@
 ,runtime,num_features,features_per_second,normalized_runtime,normalized_runtime_per_feature,rsquared,rmse,mae
-getML: FastProp,0 days 00:00:00.434939,134,308.07147258163894,1.0,1.0,0.9950584802720546,0.7236738045405348,0.5515934724757293
-featuretools,0 days 00:14:36.193549,158,0.1803254477616112,2014.5205396618837,1708.4192852741837,0.9948315985902971,0.7484233883213018,0.571029394526552
-tsfresh,0 days 00:00:31.641862,120,3.7924469626292274,72.75011438385613,81.23290203327171,0.9938361598092297,0.7906022324089085,0.5986000367767544
+getML: FastProp,0 days 00:00:00.398347,134,336.3605785401951,1.0,1.0,0.9950578985273825,0.7237156701589754,0.5516075682525006
+featuretools,0 days 00:14:21.611109,158,0.18337738975122106,2162.9662304473236,1834.2532795156408,0.9947842730879888,0.7594597450557347,0.5828310842859015
+tsfresh,0 days 00:00:26.638362,120,4.504788590271459,66.87225459210187,74.66733938782374,0.9938361598092297,0.7906022324089085,0.5986000367767544
diff --git a/fastprop_benchmark/dodgers_prop.ipynb b/fastprop_benchmark/dodgers_prop.ipynb
index f65e8bd..1125d6b 100644
--- a/fastprop_benchmark/dodgers_prop.ipynb
+++ b/fastprop_benchmark/dodgers_prop.ipynb
@@ -1,2917 +1,2916 @@
 {
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Propositionalization: Traffic near Dodgers' stadium\n",
-    "\n",
-    "In this notebook, we compare getML's FastProp against well-known feature engineering libraries featuretools and tsfresh.\n",
-    "\n",
-    "Summary:\n",
-    "\n",
-    "- Prediction type: __Regression model__\n",
-    "- Domain: __Transportation__\n",
-    "- Prediction target: __traffic volume__ \n",
-    "- Source data: __Univariate time series__\n",
-    "- Population size: __47497__"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "tags": [
-     "remove_cell_on_docs"
-    ]
-   },
-   "source": [
-    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/dodgers_prop.ipynb\">\n",
-    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
-    "</a>"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Background\n",
-    "\n",
-    "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
-    "\n",
-    "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
-    "\n",
-    "In this notebook, we use traffic data that was collected for the Glendale on ramp for the 101 North freeway in Los Angeles. For further details about the data set refer to [the full notebook](../dodgers.ipynb)."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Analysis"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "1. [Loading data](#1.-Loading-data)\n",
-    "2. [Predictive modeling](#2.-Predictive-modeling)\n",
-    "3. [Comparison](#3.-Comparison)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Let's get started with the analysis and set-up your session:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Note: you may need to restart the kernel to use updated packages.\n"
-     ]
-    }
-   ],
-   "source": [
-    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "getML API version: 1.5.0\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "import os\n",
-    "import sys\n",
-    "\n",
-    "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
-    "from pathlib import Path\n",
-    "from urllib import request\n",
-    "import getml\n",
-    "import pandas as pd\n",
-    "\n",
-    "print(f\"getML API version: {getml.__version__}\\n\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
-    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
-    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "parent = Path(os.getcwd()).parent.as_posix()\n",
-    "\n",
-    "if parent not in sys.path:\n",
-    "    sys.path.append(parent)\n",
-    "\n",
-    "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=true --home-directory=/home/alex --in-memory=true --install=false --launch-browser=true --log=false --token=token in /home/alex/.getML/getml-1.5.0-x64-community-edition-linux...\n",
-      "Launched the getML Engine. The log output will be stored in /home/alex/.getML/logs/20240912171016.log.\n"
-     ]
+    "cells": [
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "# Propositionalization: Traffic near Dodgers' stadium\n",
+                "\n",
+                "In this notebook, we compare getML's FastProp against well-known feature engineering libraries featuretools and tsfresh.\n",
+                "\n",
+                "Summary:\n",
+                "\n",
+                "- Prediction type: __Regression model__\n",
+                "- Domain: __Transportation__\n",
+                "- Prediction target: __traffic volume__ \n",
+                "- Source data: __Univariate time series__\n",
+                "- Population size: __47497__"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "tags": [
+                    "remove_cell_on_docs"
+                ]
+            },
+            "source": [
+                "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/dodgers_prop.ipynb\">\n",
+                "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+                "</a>"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Background\n",
+                "\n",
+                "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
+                "\n",
+                "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
+                "\n",
+                "In this notebook, we use traffic data that was collected for the Glendale on ramp for the 101 North freeway in Los Angeles. For further details about the data set refer to [the full notebook](../dodgers.ipynb)."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Analysis"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "1. [Loading data](#1.-Loading-data)\n",
+                "2. [Predictive modeling](#2.-Predictive-modeling)\n",
+                "3. [Comparison](#3.-Comparison)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Let's get started with the analysis and set-up your session:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 1,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Note: you may need to restart the kernel to use updated packages.\n"
+                    ]
+                }
+            ],
+            "source": [
+                "%pip install -q \"getml==1.5.0\" \"featuretools==1.31.0\" \"tsfresh==0.20.3\""
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 2,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "getML API version: 1.5.0\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "import os\n",
+                "import sys\n",
+                "\n",
+                "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
+                "from pathlib import Path\n",
+                "from urllib import request\n",
+                "import getml\n",
+                "import pandas as pd\n",
+                "\n",
+                "print(f\"getML API version: {getml.__version__}\\n\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 3,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+                "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+                "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 4,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "parent = Path(os.getcwd()).parent.as_posix()\n",
+                "\n",
+                "if parent not in sys.path:\n",
+                "    sys.path.append(parent)\n",
+                "\n",
+                "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 5,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "getML Engine is already running.\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'dodgers'</span>.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Connected to project \u001b[32m'dodgers'\u001b[0m.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
+                "getml.engine.set_project(\"dodgers\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 1. Loading data"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 1.1 Download from source\n",
+                "\n",
+                "We begin by downloading the data from the UC Irvine Machine Learning repository:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 6,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "fname = \"Dodgers.data\"\n",
+                "\n",
+                "if not os.path.exists(fname):\n",
+                "    fname, res = request.urlretrieve(\n",
+                "        \"https://archive.ics.uci.edu/ml/machine-learning-databases/event-detection/\"\n",
+                "        + fname,\n",
+                "        fname,\n",
+                "    )\n",
+                "\n",
+                "data_full_pandas = pd.read_csv(fname, header=None)\n",
+                "data_full_pandas.columns = [\"ds\", \"y\"]"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 7,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_full_pandas[\"ds\"] = pd.to_datetime(data_full_pandas[\"ds\"], format=\"%m/%d/%Y %H:%M\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 8,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>ds</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>2005-04-10 00:00:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>1</th>\n",
+                            "      <td>2005-04-10 00:05:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2</th>\n",
+                            "      <td>2005-04-10 00:10:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>3</th>\n",
+                            "      <td>2005-04-10 00:15:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>4</th>\n",
+                            "      <td>2005-04-10 00:20:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>50395</th>\n",
+                            "      <td>2005-10-01 23:35:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>50396</th>\n",
+                            "      <td>2005-10-01 23:40:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>50397</th>\n",
+                            "      <td>2005-10-01 23:45:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>50398</th>\n",
+                            "      <td>2005-10-01 23:50:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>50399</th>\n",
+                            "      <td>2005-10-01 23:55:00</td>\n",
+                            "      <td>-1</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>50400 rows × 2 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "                       ds  y\n",
+                            "0     2005-04-10 00:00:00 -1\n",
+                            "1     2005-04-10 00:05:00 -1\n",
+                            "2     2005-04-10 00:10:00 -1\n",
+                            "3     2005-04-10 00:15:00 -1\n",
+                            "4     2005-04-10 00:20:00 -1\n",
+                            "...                   ... ..\n",
+                            "50395 2005-10-01 23:35:00 -1\n",
+                            "50396 2005-10-01 23:40:00 -1\n",
+                            "50397 2005-10-01 23:45:00 -1\n",
+                            "50398 2005-10-01 23:50:00 -1\n",
+                            "50399 2005-10-01 23:55:00 -1\n",
+                            "\n",
+                            "[50400 rows x 2 columns]"
+                        ]
+                    },
+                    "execution_count": 8,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_full_pandas"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 1.2 Prepare data for getML"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 9,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_full = getml.data.DataFrame.from_pandas(data_full_pandas, \"data_full\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 10,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_full.set_role(\"y\", getml.data.roles.target)\n",
+                "data_full.set_role(\"ds\", getml.data.roles.time_stamp)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 11,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th> name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"time_stamp\">                         ds</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"target\">     y</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">                 time_stamp</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">target</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> unit</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">time stamp, comparison only</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">      </th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-04-10</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-04-10 00:05:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-04-10 00:10:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-04-10 00:15:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-04-10 00:20:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>50395</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-10-01 23:35:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>50396</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-10-01 23:40:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>50397</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-10-01 23:45:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>50398</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-10-01 23:50:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>50399</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2005-10-01 23:55:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">-1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    50400 rows x 2 columns<br />\n",
+                            "    memory usage: 0.81 MB<br />\n",
+                            "    name: data_full<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            " name                            ds        y\n",
+                            " role                    time_stamp   target\n",
+                            " unit   time stamp, comparison only         \n",
+                            "    0                    2005-04-10      -1 \n",
+                            "    1           2005-04-10 00:05:00      -1 \n",
+                            "    2           2005-04-10 00:10:00      -1 \n",
+                            "    3           2005-04-10 00:15:00      -1 \n",
+                            "    4           2005-04-10 00:20:00      -1 \n",
+                            "                                ...      ...\n",
+                            "50395           2005-10-01 23:35:00      -1 \n",
+                            "50396           2005-10-01 23:40:00      -1 \n",
+                            "50397           2005-10-01 23:45:00      -1 \n",
+                            "50398           2005-10-01 23:50:00      -1 \n",
+                            "50399           2005-10-01 23:55:00      -1 \n",
+                            "\n",
+                            "\n",
+                            "50400 rows x 2 columns\n",
+                            "memory usage: 0.81 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 11,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_full"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 12,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.numerical {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.numerical {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  td.char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th>     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th>     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>train</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td>...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    50400 rows<br />\n",
+                            "    \n",
+                            "    type: StringColumnView<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            "             \n",
+                            "    0   train\n",
+                            "    1   train\n",
+                            "    2   train\n",
+                            "    3   train\n",
+                            "    4   train\n",
+                            "        ...  \n",
+                            "\n",
+                            "\n",
+                            "50400 rows\n",
+                            "type: StringColumnView"
+                        ]
+                    },
+                    "execution_count": 12,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "split = getml.data.split.time(\n",
+                "    population=data_full, time_stamp=\"ds\", test=getml.data.time.datetime(2005, 8, 20)\n",
+                ")\n",
+                "split"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 1.3 Define relational model\n",
+                "\n",
+                "To start with relational learning, we need to specify the data model. We manually replicate the appropriate time series structure by setting time series related join conditions (`horizon`, `memory` and `allow_lagged_targets`). This is done abstractly using [Placeholders](https://docs.getml.com/latest/user_guide/data_model/data_model.html#placeholders)\n",
+                "\n",
+                "The data model consists of two tables:\n",
+                "* __Population table__ `traffic_{test/train}`: holds target and the contemporarily available time-based components\n",
+                "* __Peripheral table__ `traffic`: same table as the population table\n",
+                "* Join between both placeholders specifies (`horizon`) to prevent leaks and (`memory`) that keeps the computations feasible"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 13,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
+                            "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
+                            "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">data_full</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"30.0\" x=\"324.0\" font-size=\"7pt\" >ds &lt;= ds</tspan><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 2.0 hours</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Horizon: 1.0 hours</tspan><tspan y=\"60.0\" x=\"324.0\" font-size=\"7pt\" >Lagged targets allowed</tspan></text></svg></div>\n",
+                            "            </div>\n",
+                            "\n",
+                            "            <div style='margin-top: 15px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
+                            "            <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">data frames</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">staging table              </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DATA_FULL__STAGING_TABLE_2</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "            </div>\n",
+                            "            \n",
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
+                            "<div style='margin-top: 15px;'>\n",
+                            "<div style='float: left; margin-right: 50px;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">subset</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">rows   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "<div style='float: left;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> rows</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_full</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">50400</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DataFrame</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "data model\n",
+                            "\n",
+                            "  population:\n",
+                            "    columns:\n",
+                            "    - y: target\n",
+                            "    - ds: time_stamp\n",
+                            "\n",
+                            "    joins:\n",
+                            "    - right: 'data_full'\n",
+                            "      time_stamps: (population.ds, data_full.ds)\n",
+                            "      relationship: 'many-to-many'\n",
+                            "      memory: 7200.0\n",
+                            "      horizon: 3600.0\n",
+                            "      lagged_targets: True\n",
+                            "\n",
+                            "  data_full:\n",
+                            "    columns:\n",
+                            "    - y: target\n",
+                            "    - ds: time_stamp\n",
+                            "\n",
+                            "\n",
+                            "container\n",
+                            "\n",
+                            "  population\n",
+                            "      subset   name        rows      type\n",
+                            "  0   test     data_full   unknown   View\n",
+                            "  1   train    data_full   unknown   View\n",
+                            "\n",
+                            "  peripheral\n",
+                            "      name         rows   type     \n",
+                            "  0   data_full   50400   DataFrame"
+                        ]
+                    },
+                    "execution_count": 13,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "# 1. The horizon is 1 hour (we predict the traffic volume in one hour).\n",
+                "# 2. The memory is 2 hours, so we allow the algorithm to\n",
+                "#    use information from up to 2 hours ago.\n",
+                "# 3. We allow lagged targets. Thus, the algorithm can\n",
+                "#    identify autoregressive processes.\n",
+                "\n",
+                "time_series = getml.data.TimeSeries(\n",
+                "    population=data_full,\n",
+                "    alias=\"population\",\n",
+                "    split=split,\n",
+                "    time_stamps=\"ds\",\n",
+                "    horizon=getml.data.time.hours(1),\n",
+                "    memory=getml.data.time.hours(2),\n",
+                "    lagged_targets=True,\n",
+                ")\n",
+                "\n",
+                "time_series"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 2. Predictive modeling\n",
+                "\n",
+                "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.1 Propositionalization with getML's FastProp"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 14,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "seasonal = getml.preprocessors.Seasonal()\n",
+                "\n",
+                "fast_prop = getml.feature_learning.FastProp(\n",
+                "    loss_function=getml.feature_learning.loss_functions.SquareLoss,\n",
+                "    num_threads=1,\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "__Build the pipeline__"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 15,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=['data_full'],\n",
+                            "         predictors=[],\n",
+                            "         preprocessors=['Seasonal'],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['feature learning', 'fastprop'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=['data_full'],\n",
+                            "         predictors=[],\n",
+                            "         preprocessors=['Seasonal'],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['feature learning', 'fastprop'])"
+                        ]
+                    },
+                    "execution_count": 15,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_fl = getml.pipeline.Pipeline(\n",
+                "    preprocessors=[seasonal],\n",
+                "    feature_learners=[fast_prop],\n",
+                "    data_model=time_series.data_model,\n",
+                "    tags=[\"feature learning\", \"fastprop\"],\n",
+                ")\n",
+                "\n",
+                "pipe_fp_fl"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 16,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "pipe_fp_fl.check(time_series.train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 17,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "benchmark = Benchmark()"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 18,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Trying 526 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:08\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:08.574815.\n",
+                        "\n",
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"fastprop\"):\n",
+                "    pipe_fp_fl.fit(time_series.train)\n",
+                "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 19,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 20,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "predictor = getml.predictors.XGBoostRegressor()\n",
+                "\n",
+                "pipe_fp_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 21,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:09\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:09.859138.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])"
+                        ]
+                    },
+                    "execution_count": 21,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.fit(fastprop_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 22,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used      </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">   rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:03:14</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">5.4188</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">7.5347</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.699</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:03:15</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">5.6151</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">7.8243</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6747</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used         target       mae      rmse   rsquared\n",
+                            "0   2024-09-13 13:03:14   fastprop_train   y         5.4188    7.5347     0.699 \n",
+                            "1   2024-09-13 13:03:15   fastprop_test    y         5.6151    7.8243     0.6747"
+                        ]
+                    },
+                    "execution_count": 22,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.score(fastprop_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.2 Propositionalization with featuretools"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 23,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_train = time_series.train.population.to_df(\"data_train\")\n",
+                "data_test = time_series.test.population.to_df(\"data_test\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 24,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "dfs_pandas = {}\n",
+                "\n",
+                "for df in getml.project.data_frames:\n",
+                "    dfs_pandas[df.name] = df.to_pandas()\n",
+                "    dfs_pandas[df.name][\"id\"] = 1"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 25,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "ft_builder = FTTimeSeriesBuilder(\n",
+                "    num_features=200,\n",
+                "    horizon=pd.Timedelta(hours=1),\n",
+                "    memory=pd.Timedelta(hours=2),\n",
+                "    column_id=\"id\",\n",
+                "    time_stamp=\"ds\",\n",
+                "    target=\"y\",\n",
+                "    allow_lagged_targets=True,\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 26,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "featuretools: Trying features...\n",
+                        "Selecting the best out of 118 features...\n",
+                        "Time taken: 0h:8m:51.202135\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"featuretools\"):\n",
+                "    featuretools_train = ft_builder.fit(dfs_pandas[\"data_train\"])\n",
+                "\n",
+                "featuretools_test = ft_builder.transform(dfs_pandas[\"data_test\"])"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 27,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_featuretools_train = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_train, name=\"featuretools_train\", roles=data_train.roles\n",
+                ")\n",
+                "\n",
+                "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_test, name=\"featuretools_test\", roles=data_train.roles\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 28,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_featuretools_train.set_role(\n",
+                "    df_featuretools_train.roles.unused, getml.data.roles.numerical\n",
+                ")\n",
+                "\n",
+                "df_featuretools_test.set_role(\n",
+                "    df_featuretools_test.roles.unused, getml.data.roles.numerical\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 29,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])"
+                        ]
+                    },
+                    "execution_count": 29,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "predictor = getml.predictors.XGBoostRegressor()\n",
+                "\n",
+                "pipe_ft_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"prediction\", \"featuretools\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_ft_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 30,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">label                  </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">message                         </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">WARNING</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    type      label                     message                         \n",
+                            "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
+                        ]
+                    },
+                    "execution_count": 30,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.check(df_featuretools_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 31,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:03.563567.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])"
+                        ]
+                    },
+                    "execution_count": 31,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.fit(df_featuretools_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 32,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">   rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:15:05</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">5.4482</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">7.568</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6962</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:15:06</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">6.0863</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">8.5009</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6498</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used             target       mae      rmse   rsquared\n",
+                            "0   2024-09-13 13:15:05   featuretools_train   y         5.4482    7.568      0.6962\n",
+                            "1   2024-09-13 13:15:06   featuretools_test    y         6.0863    8.5009     0.6498"
+                        ]
+                    },
+                    "execution_count": 32,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.score(df_featuretools_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.3 Propositionalization with tsfresh"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 33,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "tsfresh_builder = TSFreshBuilder(\n",
+                "    num_features=200,\n",
+                "    horizon=20,\n",
+                "    memory=60,\n",
+                "    column_id=\"id\",\n",
+                "    time_stamp=\"ds\",\n",
+                "    target=\"y\",\n",
+                "    allow_lagged_targets=True,\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 34,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:12<00:00,  3.18it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.26it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.23it/s]\n"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Selecting the best out of 13 features...\n",
+                        "Time taken: 0h:0m:31.919565\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:03<00:00, 11.31it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:01<00:00, 20.24it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 19.32it/s]\n"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"tsfresh\"):\n",
+                "    tsfresh_train = tsfresh_builder.fit(dfs_pandas[\"data_train\"])\n",
+                "\n",
+                "tsfresh_test = tsfresh_builder.transform(dfs_pandas[\"data_test\"])"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 35,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_tsfresh_train = getml.data.DataFrame.from_pandas(\n",
+                "    tsfresh_train, name=\"tsfresh_train\", roles=data_train.roles\n",
+                ")\n",
+                "\n",
+                "df_tsfresh_test = getml.data.DataFrame.from_pandas(\n",
+                "    tsfresh_test, name=\"tsfresh_test\", roles=data_train.roles\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 36,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "df_tsfresh_train.set_role(df_tsfresh_train.roles.unused, getml.data.roles.numerical)\n",
+                "\n",
+                "df_tsfresh_test.set_role(df_tsfresh_test.roles.unused, getml.data.roles.numerical)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 37,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])"
+                        ]
+                    },
+                    "execution_count": 37,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"predicition\", \"tsfresh\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_tsf_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 38,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K⠇ XGBoost: Trained tree 95.  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━    95% • 00:01"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:01.448143.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostRegressor'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])"
+                        ]
+                    },
+                    "execution_count": 38,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.fit(df_tsfresh_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 39,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    mae</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">   rmse</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">rsquared</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:15:49</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">6.3146</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">8.2348</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.6418</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:15:49</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">6.7886</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">8.9134</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.5778</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used        target       mae      rmse   rsquared\n",
+                            "0   2024-09-13 13:15:49   tsfresh_train   y         6.3146    8.2348     0.6418\n",
+                            "1   2024-09-13 13:15:49   tsfresh_test    y         6.7886    8.9134     0.5778"
+                        ]
+                    },
+                    "execution_count": 39,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.score(df_tsfresh_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 3. Comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 40,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "num_features = dict(\n",
+                "    fastprop=526,\n",
+                "    featuretools=59,\n",
+                "    tsfresh=12,\n",
+                ")\n",
+                "\n",
+                "runtime_per_feature = [\n",
+                "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
+                "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
+                "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
+                "]\n",
+                "\n",
+                "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
+                "\n",
+                "normalized_runtime_per_feature = [\n",
+                "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
+                "]\n",
+                "\n",
+                "comparison = pd.DataFrame(\n",
+                "    dict(\n",
+                "        runtime=[\n",
+                "            benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"featuretools\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"],\n",
+                "        ],\n",
+                "        num_features=num_features.values(),\n",
+                "        features_per_second=features_per_second,\n",
+                "        normalized_runtime=[\n",
+                "            1,\n",
+                "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "        ],\n",
+                "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
+                "        rsquared=[pipe_fp_pr.rsquared, pipe_ft_pr.rsquared, pipe_tsf_pr.rsquared],\n",
+                "        rmse=[pipe_fp_pr.rmse, pipe_ft_pr.rmse, pipe_tsf_pr.rmse],\n",
+                "        mae=[pipe_fp_pr.mae, pipe_ft_pr.mae, pipe_tsf_pr.mae],\n",
+                "    )\n",
+                ")\n",
+                "\n",
+                "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 41,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>runtime</th>\n",
+                            "      <th>num_features</th>\n",
+                            "      <th>features_per_second</th>\n",
+                            "      <th>normalized_runtime</th>\n",
+                            "      <th>normalized_runtime_per_feature</th>\n",
+                            "      <th>rsquared</th>\n",
+                            "      <th>rmse</th>\n",
+                            "      <th>mae</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>getML: FastProp</th>\n",
+                            "      <td>0 days 00:00:12.106112</td>\n",
+                            "      <td>526</td>\n",
+                            "      <td>43.449924</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>0.674740</td>\n",
+                            "      <td>7.824265</td>\n",
+                            "      <td>5.615102</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>featuretools</th>\n",
+                            "      <td>0 days 00:08:51.202688</td>\n",
+                            "      <td>59</td>\n",
+                            "      <td>0.111069</td>\n",
+                            "      <td>43.878884</td>\n",
+                            "      <td>391.198566</td>\n",
+                            "      <td>0.649768</td>\n",
+                            "      <td>8.500887</td>\n",
+                            "      <td>6.086277</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>tsfresh</th>\n",
+                            "      <td>0 days 00:00:31.919755</td>\n",
+                            "      <td>12</td>\n",
+                            "      <td>0.375943</td>\n",
+                            "      <td>2.636664</td>\n",
+                            "      <td>115.575929</td>\n",
+                            "      <td>0.577811</td>\n",
+                            "      <td>8.913408</td>\n",
+                            "      <td>6.788610</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "                               runtime  num_features  features_per_second  \\\n",
+                            "getML: FastProp 0 days 00:00:12.106112           526            43.449924   \n",
+                            "featuretools    0 days 00:08:51.202688            59             0.111069   \n",
+                            "tsfresh         0 days 00:00:31.919755            12             0.375943   \n",
+                            "\n",
+                            "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
+                            "getML: FastProp            1.000000                        1.000000  0.674740   \n",
+                            "featuretools              43.878884                      391.198566  0.649768   \n",
+                            "tsfresh                    2.636664                      115.575929  0.577811   \n",
+                            "\n",
+                            "                     rmse       mae  \n",
+                            "getML: FastProp  7.824265  5.615102  \n",
+                            "featuretools     8.500887  6.086277  \n",
+                            "tsfresh          8.913408  6.788610  "
+                        ]
+                    },
+                    "execution_count": 41,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 42,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# export for further use\n",
+                "comparison.to_csv(\"comparisons/dodgers.csv\")"
+            ]
+        }
+    ],
+    "metadata": {
+        "jupytext": {
+            "encoding": "# -*- coding: utf-8 -*-",
+            "formats": "ipynb,py:percent,md"
+        },
+        "kernelspec": {
+            "display_name": "Python 3 (ipykernel)",
+            "language": "python",
+            "name": "python3"
+        },
+        "language_info": {
+            "codemirror_mode": {
+                "name": "ipython",
+                "version": 3
+            },
+            "file_extension": ".py",
+            "mimetype": "text/x-python",
+            "name": "python",
+            "nbconvert_exporter": "python",
+            "pygments_lexer": "ipython3",
+            "version": "3.11.9"
+        },
+        "toc": {
+            "base_numbering": 1,
+            "nav_menu": {},
+            "number_sections": false,
+            "sideBar": true,
+            "skip_h1_title": false,
+            "title_cell": "Table of Contents",
+            "title_sidebar": "Contents",
+            "toc_cell": false,
+            "toc_position": {},
+            "toc_section_display": true,
+            "toc_window_display": true
+        }
     },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'dodgers'</span>.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'dodgers'\u001b[0m.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
-    "getml.engine.set_project(\"dodgers\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 1. Loading data"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 1.1 Download from source\n",
-    "\n",
-    "We begin by downloading the data from the UC Irvine Machine Learning repository:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "fname = \"Dodgers.data\"\n",
-    "\n",
-    "if not os.path.exists(fname):\n",
-    "    fname, res = request.urlretrieve(\n",
-    "        \"https://archive.ics.uci.edu/ml/machine-learning-databases/event-detection/\"\n",
-    "        + fname,\n",
-    "        fname,\n",
-    "    )\n",
-    "\n",
-    "data_full_pandas = pd.read_csv(fname, header=None)\n",
-    "data_full_pandas.columns = [\"ds\", \"y\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_full_pandas[\"ds\"] = pd.to_datetime(data_full_pandas[\"ds\"], format=\"%m/%d/%Y %H:%M\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>ds</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>2005-04-10 00:00:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>2005-04-10 00:05:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>2005-04-10 00:10:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>2005-04-10 00:15:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>2005-04-10 00:20:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50395</th>\n",
-       "      <td>2005-10-01 23:35:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50396</th>\n",
-       "      <td>2005-10-01 23:40:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50397</th>\n",
-       "      <td>2005-10-01 23:45:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50398</th>\n",
-       "      <td>2005-10-01 23:50:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50399</th>\n",
-       "      <td>2005-10-01 23:55:00</td>\n",
-       "      <td>-1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>50400 rows × 2 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                       ds  y\n",
-       "0     2005-04-10 00:00:00 -1\n",
-       "1     2005-04-10 00:05:00 -1\n",
-       "2     2005-04-10 00:10:00 -1\n",
-       "3     2005-04-10 00:15:00 -1\n",
-       "4     2005-04-10 00:20:00 -1\n",
-       "...                   ... ..\n",
-       "50395 2005-10-01 23:35:00 -1\n",
-       "50396 2005-10-01 23:40:00 -1\n",
-       "50397 2005-10-01 23:45:00 -1\n",
-       "50398 2005-10-01 23:50:00 -1\n",
-       "50399 2005-10-01 23:55:00 -1\n",
-       "\n",
-       "[50400 rows x 2 columns]"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_full_pandas"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 1.2 Prepare data for getML"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_full = getml.data.DataFrame.from_pandas(data_full_pandas, \"data_full\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_full.set_role(\"y\", getml.data.roles.target)\n",
-    "data_full.set_role(\"ds\", getml.data.roles.time_stamp)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th> name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"time_stamp\">                         ds</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"target\">     y</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">                 time_stamp</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">target</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> unit</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">time stamp, comparison only</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">      </th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-04-10</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-04-10 00:05:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-04-10 00:10:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-04-10 00:15:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-04-10 00:20:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">...</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>50395</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-10-01 23:35:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>50396</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-10-01 23:40:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>50397</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-10-01 23:45:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>50398</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-10-01 23:50:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>50399</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2005-10-01 23:55:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">-1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    50400 rows x 2 columns<br />\n",
-       "    memory usage: 0.81 MB<br />\n",
-       "    name: data_full<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       " name                            ds        y\n",
-       " role                    time_stamp   target\n",
-       " unit   time stamp, comparison only         \n",
-       "    0                    2005-04-10      -1 \n",
-       "    1           2005-04-10 00:05:00      -1 \n",
-       "    2           2005-04-10 00:10:00      -1 \n",
-       "    3           2005-04-10 00:15:00      -1 \n",
-       "    4           2005-04-10 00:20:00      -1 \n",
-       "                                ...      ...\n",
-       "50395           2005-10-01 23:35:00      -1 \n",
-       "50396           2005-10-01 23:40:00      -1 \n",
-       "50397           2005-10-01 23:45:00      -1 \n",
-       "50398           2005-10-01 23:50:00      -1 \n",
-       "50399           2005-10-01 23:55:00      -1 \n",
-       "\n",
-       "\n",
-       "50400 rows x 2 columns\n",
-       "memory usage: 0.81 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 11,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_full"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.numerical {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.numerical {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  td.char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th>     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th>     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td>train</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td>...</td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    50400 rows<br />\n",
-       "    \n",
-       "    type: StringColumnView<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       "             \n",
-       "    0   train\n",
-       "    1   train\n",
-       "    2   train\n",
-       "    3   train\n",
-       "    4   train\n",
-       "        ...  \n",
-       "\n",
-       "\n",
-       "50400 rows\n",
-       "type: StringColumnView"
-      ]
-     },
-     "execution_count": 12,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "split = getml.data.split.time(\n",
-    "    population=data_full, time_stamp=\"ds\", test=getml.data.time.datetime(2005, 8, 20)\n",
-    ")\n",
-    "split"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 1.3 Define relational model\n",
-    "\n",
-    "To start with relational learning, we need to specify the data model. We manually replicate the appropriate time series structure by setting time series related join conditions (`horizon`, `memory` and `allow_lagged_targets`). This is done abstractly using [Placeholders](https://docs.getml.com/latest/user_guide/data_model/data_model.html#placeholders)\n",
-    "\n",
-    "The data model consists of two tables:\n",
-    "* __Population table__ `traffic_{test/train}`: holds target and the contemporarily available time-based components\n",
-    "* __Peripheral table__ `traffic`: same table as the population table\n",
-    "* Join between both placeholders specifies (`horizon`) to prevent leaks and (`memory`) that keeps the computations feasible"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
-       "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
-       "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">data_full</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"30.0\" x=\"324.0\" font-size=\"7pt\" >ds &lt;= ds</tspan><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 2.0 hours</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Horizon: 1.0 hours</tspan><tspan y=\"60.0\" x=\"324.0\" font-size=\"7pt\" >Lagged targets allowed</tspan></text></svg></div>\n",
-       "            </div>\n",
-       "\n",
-       "            <div style='margin-top: 15px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
-       "            <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">data frames</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">staging table              </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DATA_FULL__STAGING_TABLE_2</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "            </div>\n",
-       "            \n",
-       "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
-       "<div style='margin-top: 15px;'>\n",
-       "<div style='float: left; margin-right: 50px;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">subset</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">rows   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "<div style='float: left;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> rows</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_full</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">50400</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DataFrame</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "data model\n",
-       "\n",
-       "  population:\n",
-       "    columns:\n",
-       "    - y: target\n",
-       "    - ds: time_stamp\n",
-       "\n",
-       "    joins:\n",
-       "    - right: 'data_full'\n",
-       "      time_stamps: (population.ds, data_full.ds)\n",
-       "      relationship: 'many-to-many'\n",
-       "      memory: 7200.0\n",
-       "      horizon: 3600.0\n",
-       "      lagged_targets: True\n",
-       "\n",
-       "  data_full:\n",
-       "    columns:\n",
-       "    - y: target\n",
-       "    - ds: time_stamp\n",
-       "\n",
-       "\n",
-       "container\n",
-       "\n",
-       "  population\n",
-       "      subset   name        rows      type\n",
-       "  0   test     data_full   unknown   View\n",
-       "  1   train    data_full   unknown   View\n",
-       "\n",
-       "  peripheral\n",
-       "      name         rows   type     \n",
-       "  0   data_full   50400   DataFrame"
-      ]
-     },
-     "execution_count": 13,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# 1. The horizon is 1 hour (we predict the traffic volume in one hour).\n",
-    "# 2. The memory is 2 hours, so we allow the algorithm to\n",
-    "#    use information from up to 2 hours ago.\n",
-    "# 3. We allow lagged targets. Thus, the algorithm can\n",
-    "#    identify autoregressive processes.\n",
-    "\n",
-    "time_series = getml.data.TimeSeries(\n",
-    "    population=data_full,\n",
-    "    alias=\"population\",\n",
-    "    split=split,\n",
-    "    time_stamps=\"ds\",\n",
-    "    horizon=getml.data.time.hours(1),\n",
-    "    memory=getml.data.time.hours(2),\n",
-    "    lagged_targets=True,\n",
-    ")\n",
-    "\n",
-    "time_series"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 2. Predictive modeling\n",
-    "\n",
-    "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.1 Propositionalization with getML's FastProp"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "seasonal = getml.preprocessors.Seasonal()\n",
-    "\n",
-    "fast_prop = getml.feature_learning.FastProp(\n",
-    "    loss_function=getml.feature_learning.loss_functions.SquareLoss,\n",
-    "    num_threads=1,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "__Build the pipeline__"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=['data_full'],\n",
-       "         predictors=[],\n",
-       "         preprocessors=['Seasonal'],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['feature learning', 'fastprop'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=['data_full'],\n",
-       "         predictors=[],\n",
-       "         preprocessors=['Seasonal'],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['feature learning', 'fastprop'])"
-      ]
-     },
-     "execution_count": 15,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_fl = getml.pipeline.Pipeline(\n",
-    "    preprocessors=[seasonal],\n",
-    "    feature_learners=[fast_prop],\n",
-    "    data_model=time_series.data_model,\n",
-    "    tags=[\"feature learning\", \"fastprop\"],\n",
-    ")\n",
-    "\n",
-    "pipe_fp_fl"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pipe_fp_fl.check(time_series.train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "benchmark = Benchmark()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Trying 526 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:08\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:08.380775.\n",
-      "\n",
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"fastprop\"):\n",
-    "    pipe_fp_fl.fit(time_series.train)\n",
-    "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "predictor = getml.predictors.XGBoostRegressor()\n",
-    "\n",
-    "pipe_fp_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:10\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:10.075707.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])"
-      ]
-     },
-     "execution_count": 21,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.fit(fastprop_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used      </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">   rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:10:43</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">5.4188</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">7.5347</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.699</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:10:43</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">5.6151</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">7.8243</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6747</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:10:43   fastprop_train   y         5.4188    7.5347     0.699 \n",
-       "1   2024-09-12 17:10:43   fastprop_test    y         5.6151    7.8243     0.6747"
-      ]
-     },
-     "execution_count": 22,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.score(fastprop_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.2 Propositionalization with featuretools"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_train = time_series.train.population.to_df(\"data_train\")\n",
-    "data_test = time_series.test.population.to_df(\"data_test\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "dfs_pandas = {}\n",
-    "\n",
-    "for df in getml.project.data_frames:\n",
-    "    dfs_pandas[df.name] = df.to_pandas()\n",
-    "    dfs_pandas[df.name][\"id\"] = 1"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "ft_builder = FTTimeSeriesBuilder(\n",
-    "    num_features=200,\n",
-    "    horizon=pd.Timedelta(hours=1),\n",
-    "    memory=pd.Timedelta(hours=2),\n",
-    "    column_id=\"id\",\n",
-    "    time_stamp=\"ds\",\n",
-    "    target=\"y\",\n",
-    "    allow_lagged_targets=True,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "featuretools: Trying features...\n",
-      "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:8m:56.480767\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"featuretools\"):\n",
-    "    featuretools_train = ft_builder.fit(dfs_pandas[\"data_train\"])\n",
-    "\n",
-    "featuretools_test = ft_builder.transform(dfs_pandas[\"data_test\"])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_featuretools_train = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_train, name=\"featuretools_train\", roles=data_train.roles\n",
-    ")\n",
-    "\n",
-    "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_test, name=\"featuretools_test\", roles=data_train.roles\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 28,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_featuretools_train.set_role(\n",
-    "    df_featuretools_train.roles.unused, getml.data.roles.numerical\n",
-    ")\n",
-    "\n",
-    "df_featuretools_test.set_role(\n",
-    "    df_featuretools_test.roles.unused, getml.data.roles.numerical\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 29,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])"
-      ]
-     },
-     "execution_count": 29,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "predictor = getml.predictors.XGBoostRegressor()\n",
-    "\n",
-    "pipe_ft_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"prediction\", \"featuretools\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_ft_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">label                  </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">message                         </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">WARNING</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    type      label                     message                         \n",
-       "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
-      ]
-     },
-     "execution_count": 30,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.check(df_featuretools_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 31,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:03.540051.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])"
-      ]
-     },
-     "execution_count": 31,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.fit(df_featuretools_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">   rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:22:38</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">5.4482</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">7.568</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6962</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:22:38</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">6.0863</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">8.5009</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6498</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:22:38   featuretools_train   y         5.4482    7.568      0.6962\n",
-       "1   2024-09-12 17:22:38   featuretools_test    y         6.0863    8.5009     0.6498"
-      ]
-     },
-     "execution_count": 32,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.score(df_featuretools_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.3 Propositionalization with tsfresh"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "tsfresh_builder = TSFreshBuilder(\n",
-    "    num_features=200,\n",
-    "    horizon=20,\n",
-    "    memory=60,\n",
-    "    column_id=\"id\",\n",
-    "    time_stamp=\"ds\",\n",
-    "    target=\"y\",\n",
-    "    allow_lagged_targets=True,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 34,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:12<00:00,  3.14it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.17it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.30it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Selecting the best out of 13 features...\n",
-      "Time taken: 0h:0m:32.318403\n",
-      "\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 10.97it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 18.81it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:02<00:00, 19.18it/s]\n"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"tsfresh\"):\n",
-    "    tsfresh_train = tsfresh_builder.fit(dfs_pandas[\"data_train\"])\n",
-    "\n",
-    "tsfresh_test = tsfresh_builder.transform(dfs_pandas[\"data_test\"])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 35,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_tsfresh_train = getml.data.DataFrame.from_pandas(\n",
-    "    tsfresh_train, name=\"tsfresh_train\", roles=data_train.roles\n",
-    ")\n",
-    "\n",
-    "df_tsfresh_test = getml.data.DataFrame.from_pandas(\n",
-    "    tsfresh_test, name=\"tsfresh_test\", roles=data_train.roles\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 36,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df_tsfresh_train.set_role(df_tsfresh_train.roles.unused, getml.data.roles.numerical)\n",
-    "\n",
-    "df_tsfresh_test.set_role(df_tsfresh_test.roles.unused, getml.data.roles.numerical)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 37,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])"
-      ]
-     },
-     "execution_count": 37,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"predicition\", \"tsfresh\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_tsf_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 38,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K⠸ XGBoost: Trained tree 99.  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸  99% • 00:01"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:01.500237.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostRegressor'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])"
-      ]
-     },
-     "execution_count": 38,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.fit(df_tsfresh_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 39,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    mae</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">   rmse</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">rsquared</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:23:23</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">6.3146</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">8.2348</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.6418</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:23:23</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">6.7886</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">8.9134</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.5778</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:23:23   tsfresh_train   y         6.3146    8.2348     0.6418\n",
-       "1   2024-09-12 17:23:23   tsfresh_test    y         6.7886    8.9134     0.5778"
-      ]
-     },
-     "execution_count": 39,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.score(df_tsfresh_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 3. Comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 40,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "num_features = dict(\n",
-    "    fastprop=526,\n",
-    "    featuretools=59,\n",
-    "    tsfresh=12,\n",
-    ")\n",
-    "\n",
-    "runtime_per_feature = [\n",
-    "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
-    "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
-    "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
-    "]\n",
-    "\n",
-    "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
-    "\n",
-    "normalized_runtime_per_feature = [\n",
-    "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
-    "]\n",
-    "\n",
-    "comparison = pd.DataFrame(\n",
-    "    dict(\n",
-    "        runtime=[\n",
-    "            benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"featuretools\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"],\n",
-    "        ],\n",
-    "        num_features=num_features.values(),\n",
-    "        features_per_second=features_per_second,\n",
-    "        normalized_runtime=[\n",
-    "            1,\n",
-    "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "        ],\n",
-    "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
-    "        rsquared=[pipe_fp_pr.rsquared, pipe_ft_pr.rsquared, pipe_tsf_pr.rsquared],\n",
-    "        rmse=[pipe_fp_pr.rmse, pipe_ft_pr.rmse, pipe_tsf_pr.rmse],\n",
-    "        mae=[pipe_fp_pr.mae, pipe_ft_pr.mae, pipe_tsf_pr.mae],\n",
-    "    )\n",
-    ")\n",
-    "\n",
-    "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 41,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>runtime</th>\n",
-       "      <th>num_features</th>\n",
-       "      <th>features_per_second</th>\n",
-       "      <th>normalized_runtime</th>\n",
-       "      <th>normalized_runtime_per_feature</th>\n",
-       "      <th>rsquared</th>\n",
-       "      <th>rmse</th>\n",
-       "      <th>mae</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:12.036174</td>\n",
-       "      <td>526</td>\n",
-       "      <td>43.702474</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>0.674740</td>\n",
-       "      <td>7.824273</td>\n",
-       "      <td>5.615138</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>featuretools</th>\n",
-       "      <td>0 days 00:08:56.482944</td>\n",
-       "      <td>59</td>\n",
-       "      <td>0.109976</td>\n",
-       "      <td>44.572548</td>\n",
-       "      <td>397.383577</td>\n",
-       "      <td>0.649768</td>\n",
-       "      <td>8.500887</td>\n",
-       "      <td>6.086277</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:32.318518</td>\n",
-       "      <td>12</td>\n",
-       "      <td>0.371304</td>\n",
-       "      <td>2.685116</td>\n",
-       "      <td>117.699939</td>\n",
-       "      <td>0.577811</td>\n",
-       "      <td>8.913408</td>\n",
-       "      <td>6.788610</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:12.036174           526            43.702474   \n",
-       "featuretools    0 days 00:08:56.482944            59             0.109976   \n",
-       "tsfresh         0 days 00:00:32.318518            12             0.371304   \n",
-       "\n",
-       "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
-       "getML: FastProp            1.000000                        1.000000  0.674740   \n",
-       "featuretools              44.572548                      397.383577  0.649768   \n",
-       "tsfresh                    2.685116                      117.699939  0.577811   \n",
-       "\n",
-       "                     rmse       mae  \n",
-       "getML: FastProp  7.824273  5.615138  \n",
-       "featuretools     8.500887  6.086277  \n",
-       "tsfresh          8.913408  6.788610  "
-      ]
-     },
-     "execution_count": 41,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 42,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# export for further use\n",
-    "comparison.to_csv(\"comparisons/dodgers.csv\")"
-   ]
-  }
- ],
- "metadata": {
-  "jupytext": {
-   "encoding": "# -*- coding: utf-8 -*-",
-   "formats": "ipynb,py:percent,md"
-  },
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.9"
-  },
-  "toc": {
-   "base_numbering": 1,
-   "nav_menu": {},
-   "number_sections": false,
-   "sideBar": true,
-   "skip_h1_title": false,
-   "title_cell": "Table of Contents",
-   "title_sidebar": "Contents",
-   "toc_cell": false,
-   "toc_position": {},
-   "toc_section_display": true,
-   "toc_window_display": true
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
+    "nbformat": 4,
+    "nbformat_minor": 4
 }
diff --git a/fastprop_benchmark/interstate94_prop.ipynb b/fastprop_benchmark/interstate94_prop.ipynb
index 1cd414c..95b9adc 100644
--- a/fastprop_benchmark/interstate94_prop.ipynb
+++ b/fastprop_benchmark/interstate94_prop.ipynb
@@ -80,7 +80,7 @@
     }
    ],
    "source": [
-    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\""
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.31.0\""
    ]
   },
   {
@@ -119,9 +119,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "getML Engine is already running.\n",
-      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
+      "getML Engine is already running.\n"
      ]
     },
     {
@@ -1395,6 +1393,13 @@
      "metadata": {},
      "output_type": "display_data"
     },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
@@ -1478,7 +1483,7 @@
      "text": [
       "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Trying 365 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
+      "\u001b[2K  FastProp: Trying 365 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
       "\u001b[?25h"
      ]
     },
@@ -1499,7 +1504,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:03.061732.\n",
+      "Time taken: 0:00:03.058378.\n",
       "\n",
       "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
@@ -1613,7 +1618,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:05.329233.\n",
+      "Time taken: 0:00:05.192145.\n",
       "\n"
      ]
     },
@@ -1736,7 +1741,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:36:45</td>\n",
+       "              <td class=\"datetime\">2024-09-13 13:17:10</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1765,7 +1770,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:36:46</td>\n",
+       "              <td class=\"datetime\">2024-09-13 13:17:10</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -1795,8 +1800,8 @@
       ],
       "text/plain": [
        "    date time             set used         target                 mae        rmse   rsquared\n",
-       "0   2024-09-12 16:36:45   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
-       "1   2024-09-12 16:36:46   fastprop_test    traffic_volume    180.4867    261.9389     0.9827"
+       "0   2024-09-13 13:17:10   fastprop_train   traffic_volume    198.9482    292.2493     0.9779\n",
+       "1   2024-09-13 13:17:10   fastprop_test    traffic_volume    180.4867    261.9389     0.9827"
       ]
      },
      "execution_count": 19,
@@ -1868,7 +1873,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 118 features...\n",
-      "Time taken: 0h:4m:33.369809\n",
+      "Time taken: 0h:4m:27.008254\n",
       "\n"
      ]
     }
@@ -2052,7 +2057,7 @@
      "output_type": "stream",
      "text": [
       "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:02\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
       "\u001b[?25h"
      ]
     },
@@ -2073,7 +2078,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:02.103350.\n",
+      "Time taken: 0:00:01.955919.\n",
       "\n"
      ]
     },
@@ -2196,7 +2201,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:42:31</td>\n",
+       "              <td class=\"datetime\">2024-09-13 13:22:48</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2225,7 +2230,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:42:31</td>\n",
+       "              <td class=\"datetime\">2024-09-13 13:22:48</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -2255,8 +2260,8 @@
       ],
       "text/plain": [
        "    date time             set used             target                 mae        rmse   rsquared\n",
-       "0   2024-09-12 16:42:31   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
-       "1   2024-09-12 16:42:31   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
+       "0   2024-09-13 13:22:48   featuretools_train   traffic_volume    220.4023    321.1657     0.9734\n",
+       "1   2024-09-13 13:22:48   featuretools_test    traffic_volume    210.1988    317.52       0.9746"
       ]
      },
      "execution_count": 29,
@@ -2364,10 +2369,10 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:04.800454</td>\n",
+       "      <td>0 days 00:00:04.806504</td>\n",
        "      <td>461</td>\n",
-       "      <td>96.033804</td>\n",
-       "      <td>1.00000</td>\n",
+       "      <td>95.914061</td>\n",
+       "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.982678</td>\n",
        "      <td>261.938873</td>\n",
@@ -2375,11 +2380,11 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:04:33.370925</td>\n",
+       "      <td>0 days 00:04:27.009351</td>\n",
        "      <td>59</td>\n",
-       "      <td>0.215824</td>\n",
-       "      <td>56.94689</td>\n",
-       "      <td>444.963603</td>\n",
+       "      <td>0.220966</td>\n",
+       "      <td>55.551676</td>\n",
+       "      <td>434.066948</td>\n",
        "      <td>0.974582</td>\n",
        "      <td>317.519976</td>\n",
        "      <td>210.198793</td>\n",
@@ -2390,12 +2395,12 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:04.800454           461            96.033804   \n",
-       "featuretools    0 days 00:04:33.370925            59             0.215824   \n",
+       "getML: FastProp 0 days 00:00:04.806504           461            95.914061   \n",
+       "featuretools    0 days 00:04:27.009351            59             0.220966   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
-       "getML: FastProp             1.00000                        1.000000  0.982678   \n",
-       "featuretools               56.94689                      444.963603  0.974582   \n",
+       "getML: FastProp            1.000000                        1.000000  0.982678   \n",
+       "featuretools              55.551676                      434.066948  0.974582   \n",
        "\n",
        "                       rmse         mae  \n",
        "getML: FastProp  261.938873  180.486734  \n",
diff --git a/fastprop_benchmark/occupancy_prop.ipynb b/fastprop_benchmark/occupancy_prop.ipynb
index ea023dc..7be1d37 100644
--- a/fastprop_benchmark/occupancy_prop.ipynb
+++ b/fastprop_benchmark/occupancy_prop.ipynb
@@ -1,3553 +1,3565 @@
 {
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
+    "cells": [
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "slideshow": {
+                    "slide_type": "slide"
+                },
+                "tags": [
+                    "hide_input"
+                ]
+            },
+            "source": [
+                "# Propositionalization: Occupancy detection\n",
+                "\n",
+                "In this notebbok, we compare getML's FastProp against well-known feature engineering libraries featuretools and tsfresh.\n",
+                "\n",
+                "Summary:\n",
+                "\n",
+                "- Prediction type: __Binary classification__\n",
+                "- Domain: __Energy__\n",
+                "- Prediction target: __Room occupancy__\n",
+                "- Source data: __1 table, 32k rows__\n",
+                "- Population size: __32k__"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "tags": [
+                    "remove_cell_on_docs"
+                ]
+            },
+            "source": [
+                "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/occupancy_prop.ipynb\">\n",
+                "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+                "</a>"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "slideshow": {
+                    "slide_type": "slide"
+                }
+            },
+            "source": [
+                "## Background\n",
+                "\n",
+                "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
+                "\n",
+                "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
+                "\n",
+                "Our use case here is a public domain data set for predicting room occupancy from sensor data. For further details about the data set refer to [the full notebook](../occupancy.ipynb)."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Analysis"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "1. [Loading data](#1.-Loading-data)\n",
+                "2. [Predictive modeling](#2.-Predictive-modeling)\n",
+                "3. [Comparison](#3.-Comparison)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Let's get started with the analysis and set-up your session:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 1,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Note: you may need to restart the kernel to use updated packages.\n"
+                    ]
+                }
+            ],
+            "source": [
+                "%pip install -q \"getml==1.5.0\" \"featuretools==1.31.0\" \"tsfresh==0.20.3\""
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 2,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "getML API version: 1.5.0\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "import os\n",
+                "import sys\n",
+                "\n",
+                "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
+                "from pathlib import Path\n",
+                "\n",
+                "import pandas as pd\n",
+                "import getml\n",
+                "\n",
+                "print(f\"getML API version: {getml.__version__}\\n\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 3,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "getML Engine is already running.\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'occupancy'</span>.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Connected to project \u001b[32m'occupancy'\u001b[0m.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
+                "getml.engine.set_project(\"occupancy\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 4,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# If we are in Colab, we need to fetch the utils folder from the repository\n",
+                "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
+                "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 5,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "parent = Path(os.getcwd()).parent.as_posix()\n",
+                "\n",
+                "if parent not in sys.path:\n",
+                "    sys.path.append(parent)\n",
+                "\n",
+                "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 1. Loading data\n"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The data set can be downloaded directly from GitHub. It is conveniently separated into a train, a validation and a testing set. This allows us to directly benchmark our results against the results of the original paper later."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 6,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Downloading population_train... ━━━━━━━━━━━━━━━━ 100% • 554.6/554.6 kB • 00:00\n",
+                        "\u001b[2K  Downloading population_test... ━━━━━━━━━━━━━━━━━ 100% • 668.3/668.3 kB • 00:00\n",
+                        "\u001b[2K  Downloading population_validation... ━━━━━━━━━━━━ 100% • 186.5/186.5   • 00:00\n",
+                        "\u001b[2K\u001b[1A\u001b[2K  Downloading population_validation... ━━━━━━━━━━━━ 100% • 186.5/186.5   • 00:00\n",
+                        "\u001b[2K\u001b[1A\u001b[2K  Downloading population_validation... ━━━━━━━━━━━━ 100% • 186.5/186.5   • 00:00\n",
+                        "                                                           kB                   \n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "data_test, data_train, data_validate = getml.datasets.load_occupancy(roles=True)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 7,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_all, split = getml.data.split.concat(\n",
+                "    \"data_all\",\n",
+                "    train=data_train,\n",
+                "    validation=data_validate,\n",
+                "    test=data_test,\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The train set looks like this:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 8,
+            "metadata": {
+                "tags": [
+                    "hide_input"
+                ]
+            },
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th>name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"time_stamp\">               date</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"target\">Occupancy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">Temperature</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\"> Humidity</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">    Light</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">      CO2</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">HumidityRatio</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\">role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">         time_stamp</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">   target</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">  numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">    numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\">unit</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header time_stamp\">         time stamp</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">           </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">         </th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">             </th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-11 14:48:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.76</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">31</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.1333</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">437</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.3333</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1029</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.6667</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.005021</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-11 14:49:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.79</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">31</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">437</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.3333</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1000</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.005009</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-11 14:50:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.7675</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">31</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.1225</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">434</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1003</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.75</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.005022</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-11 14:51:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.7675</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">31</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.1225</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">439</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1009</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.5</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.005022</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-11 14:51:59</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.79</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">31</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.1333</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">437</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.3333</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1005</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.6667</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.00503</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>9747</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-18 09:15:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">20</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.815</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">27</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.7175</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">429</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.75</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1505</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.25</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.004213</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>9748</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-18 09:16:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">20</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.865</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">27</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.745</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">423</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.5</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1514</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.5</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.00423</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>9749</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-18 09:16:59</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">20</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.89</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">27</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.745</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">423</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.5</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1521</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.5</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.004237</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>9750</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-18 09:17:59</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">20</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.89</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">28</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.0225</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">418</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.75</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1632</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.004279</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>9751</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"time_stamp\">2015-02-18 09:19:00</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">21</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">28</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >.1</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">409</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">1864</span\n",
+                            "              ><span class=\"right\" style=\"width: 4ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "            \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 6ch\"\n",
+                            "                >.004321</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    9752 rows x 7 columns<br />\n",
+                            "    memory usage: 0.55 MB<br />\n",
+                            "    name: population_test<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            "name                  date   Occupancy   Temperature    Humidity       Light         CO2   HumidityRatio\n",
+                            "role            time_stamp      target     numerical   numerical   numerical   numerical       numerical\n",
+                            "unit            time stamp                                                                              \n",
+                            "   0   2015-02-11 14:48:00           1       21.76       31.1333    437.3333   1029.6667        0.005021\n",
+                            "   1   2015-02-11 14:49:00           1       21.79       31         437.3333   1000             0.005009\n",
+                            "   2   2015-02-11 14:50:00           1       21.7675     31.1225    434        1003.75          0.005022\n",
+                            "   3   2015-02-11 14:51:00           1       21.7675     31.1225    439        1009.5           0.005022\n",
+                            "   4   2015-02-11 14:51:59           1       21.79       31.1333    437.3333   1005.6667        0.00503 \n",
+                            "                       ...         ...       ...         ...         ...         ...           ...      \n",
+                            "9747   2015-02-18 09:15:00           1       20.815      27.7175    429.75     1505.25          0.004213\n",
+                            "9748   2015-02-18 09:16:00           1       20.865      27.745     423.5      1514.5           0.00423 \n",
+                            "9749   2015-02-18 09:16:59           1       20.89       27.745     423.5      1521.5           0.004237\n",
+                            "9750   2015-02-18 09:17:59           1       20.89       28.0225    418.75     1632             0.004279\n",
+                            "9751   2015-02-18 09:19:00           1       21          28.1       409        1864             0.004321\n",
+                            "\n",
+                            "\n",
+                            "9752 rows x 7 columns\n",
+                            "memory usage: 0.55 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 8,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_train"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### 2. Predictive modeling\n",
+                "\n",
+                "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.1 Propositionalization with getML's FastProp"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We use all possible aggregations. Because tsfresh and featuretools are single-threaded, we limit our FastProp algorithm to one thread as well, to ensure a fair comparison."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 9,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
+                            "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
+                            "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">data_all</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >date &lt;= date</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 15.0 minutes</tspan></text></svg></div>\n",
+                            "            </div>\n",
+                            "\n",
+                            "            <div style='margin-top: 15px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
+                            "            <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">data frames</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">staging table              </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_all</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DATA_ALL__STAGING_TABLE_2</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "            </div>\n",
+                            "            \n",
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
+                            "<div style='margin-top: 15px;'>\n",
+                            "<div style='float: left; margin-right: 50px;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">subset    </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name    </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">rows   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_all</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_all</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">validation</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_all</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">unknown</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "<div style='float: left;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name    </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> rows</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">data_all</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">20560</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DataFrame</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "data model\n",
+                            "\n",
+                            "  population:\n",
+                            "    columns:\n",
+                            "    - Temperature: numerical\n",
+                            "    - Humidity: numerical\n",
+                            "    - Light: numerical\n",
+                            "    - CO2: numerical\n",
+                            "    - HumidityRatio: numerical\n",
+                            "    - ...\n",
+                            "\n",
+                            "    joins:\n",
+                            "    - right: 'data_all'\n",
+                            "      time_stamps: (population.date, data_all.date)\n",
+                            "      relationship: 'many-to-many'\n",
+                            "      memory: 900.0\n",
+                            "      horizon: 0.0\n",
+                            "      lagged_targets: False\n",
+                            "\n",
+                            "  data_all:\n",
+                            "    columns:\n",
+                            "    - Temperature: numerical\n",
+                            "    - Humidity: numerical\n",
+                            "    - Light: numerical\n",
+                            "    - CO2: numerical\n",
+                            "    - HumidityRatio: numerical\n",
+                            "    - ...\n",
+                            "\n",
+                            "\n",
+                            "container\n",
+                            "\n",
+                            "  population\n",
+                            "      subset       name       rows      type\n",
+                            "  0   test         data_all   unknown   View\n",
+                            "  1   train        data_all   unknown   View\n",
+                            "  2   validation   data_all   unknown   View\n",
+                            "\n",
+                            "  peripheral\n",
+                            "      name        rows   type     \n",
+                            "  0   data_all   20560   DataFrame"
+                        ]
+                    },
+                    "execution_count": 9,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "# Our forecast horizon is 0.\n",
+                "# We do not predict the future, instead we infer\n",
+                "# the present state from current and past sensor data.\n",
+                "horizon = 0.0\n",
+                "\n",
+                "# We do not allow the time series features\n",
+                "# to use target values from the past.\n",
+                "# (Otherwise, we would need the horizon to\n",
+                "# be greater than 0.0).\n",
+                "allow_lagged_targets = False\n",
+                "\n",
+                "# We want our time series features to only use\n",
+                "# data from the last 15 minutes\n",
+                "memory = getml.data.time.minutes(15)\n",
+                "\n",
+                "time_series = getml.data.TimeSeries(\n",
+                "    population=data_all,\n",
+                "    split=split,\n",
+                "    time_stamps=\"date\",\n",
+                "    horizon=horizon,\n",
+                "    memory=memory,\n",
+                "    lagged_targets=allow_lagged_targets,\n",
+                ")\n",
+                "\n",
+                "time_series"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 10,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "feature_learner = getml.feature_learning.FastProp(\n",
+                "    loss_function=getml.feature_learning.loss_functions.CrossEntropyLoss,\n",
+                "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
+                "    num_threads=1,\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Next, we create the pipeline. In contrast to our usual approach, we create _two pipelines_ in\n",
+                "this notebook. One for feature learning (suffix `_fl`) and one for predicition (suffix `_pr`).\n",
+                "This allows for a fair comparison of runtimes."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 11,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "pipe_fp_fl = getml.pipeline.Pipeline(\n",
+                "    feature_learners=[feature_learner],\n",
+                "    data_model=time_series.data_model,\n",
+                "    tags=[\"feature learning\", \"fastprop\"],\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 12,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "pipe_fp_fl.check(time_series.train)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The wrappers around featuretools and tsfresh fit on the training set and then return the training features. We therefore measure the time it takes getML's FastProp algorithm to fit on the training set and create the training features."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 13,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "benchmark = Benchmark()"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 14,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:01.031077.\n",
+                        "\n",
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"fastprop\"):\n",
+                "    pipe_fp_fl.fit(time_series.train)\n",
+                "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 15,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                }
+            ],
+            "source": [
+                "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now we create a dedicated prediction pipeline and provide the fast prop features\n",
+                "(contrained in `fastprop_train` and `fastprop_test`.)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 16,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "predictor = getml.predictors.XGBoostClassifier()\n",
+                "\n",
+                "pipe_fp_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 17,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "OK.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:04.947655.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'fastprop'])"
+                        ]
+                    },
+                    "execution_count": 17,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.check(fastprop_train)\n",
+                "\n",
+                "pipe_fp_pr.fit(fastprop_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 18,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used      </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">accuracy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    auc</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">cross entropy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:38:49</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9995</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">1.</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.004464</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:38:49</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">fastprop_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9888</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9982</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.044213</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used         target      accuracy       auc   cross entropy\n",
+                            "0   2024-09-13 13:38:49   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
+                            "1   2024-09-13 13:38:49   fastprop_test    Occupancy     0.9888    0.9982        0.044213"
+                        ]
+                    },
+                    "execution_count": 18,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_fp_pr.score(fastprop_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.2 Propositionalization with featuretools"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 19,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data_train = time_series.train.population.to_df(\"train\")\n",
+                "data_test = time_series.test.population.to_df(\"test\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 20,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "dfs_pandas = {}\n",
+                "\n",
+                "for df in getml.project.data_frames:\n",
+                "    dfs_pandas[df.name] = df.to_pandas()\n",
+                "    dfs_pandas[df.name][\"id\"] = 1"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 21,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "ft_builder = FTTimeSeriesBuilder(\n",
+                "    num_features=200,\n",
+                "    horizon=pd.Timedelta(minutes=0),\n",
+                "    memory=pd.Timedelta(minutes=15),\n",
+                "    column_id=\"id\",\n",
+                "    time_stamp=\"date\",\n",
+                "    target=\"Occupancy\",\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The `FTTimeSeriesBuilder` provides a `fit` method that is designed to be equivilant to\n",
+                "to the `fit` method of the predictorless getML pipeline above."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 22,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "featuretools: Trying features...\n",
+                        "Selecting the best out of 262 features...\n",
+                        "Time taken: 0h:7m:20.109537\n",
+                        "\n"
+                    ]
+                }
+            ],
+            "source": [
+                "with benchmark(\"featuretools\"):\n",
+                "    featuretools_train = ft_builder.fit(dfs_pandas[\"train\"])\n",
+                "\n",
+                "featuretools_test = ft_builder.transform(dfs_pandas[\"test\"])\n",
+                "\n",
+                "df_featuretools_train = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_train, name=\"featuretools_train\", roles=data_train.roles\n",
+                ")\n",
+                "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
+                "    featuretools_test, name=\"featuretools_test\", roles=data_train.roles\n",
+                ")\n",
+                "\n",
+                "df_featuretools_train.set_role(\n",
+                "    df_featuretools_train.roles.unused, getml.data.roles.numerical\n",
+                ")\n",
+                "\n",
+                "df_featuretools_test.set_role(\n",
+                "    df_featuretools_test.roles.unused, getml.data.roles.numerical\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 23,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])"
+                        ]
+                    },
+                    "execution_count": 23,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "predictor = getml.predictors.XGBoostClassifier()\n",
+                "\n",
+                "pipe_ft_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"prediction\", \"featuretools\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_ft_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 24,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">label                  </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">message                         </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">WARNING</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    type      label                     message                         \n",
+                            "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
+                        ]
+                    },
+                    "execution_count": 24,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.check(df_featuretools_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 25,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:03.236058.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['prediction', 'featuretools'])"
+                        ]
+                    },
+                    "execution_count": 25,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.fit(df_featuretools_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 26,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">accuracy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    auc</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">cross entropy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:52:31</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9995</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">1.</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.005065</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:52:31</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">featuretools_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9885</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9972</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.049236</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used             target      accuracy       auc   cross entropy\n",
+                            "0   2024-09-13 13:52:31   featuretools_train   Occupancy     0.9995    1.            0.005065\n",
+                            "1   2024-09-13 13:52:31   featuretools_test    Occupancy     0.9885    0.9972        0.049236"
+                        ]
+                    },
+                    "execution_count": 26,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_ft_pr.score(df_featuretools_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "#### 2.3 Propositionalization with tsfresh"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 27,
+            "metadata": {
+                "lines_to_next_cell": 2
+            },
+            "outputs": [
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:02<00:00, 16.43it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.81it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.43it/s]\n"
+                    ]
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Selecting the best out of 65 features...\n",
+                        "Time taken: 0h:0m:14.295165\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "name": "stderr",
+                    "output_type": "stream",
+                    "text": [
+                        "Rolling: 100%|██████████| 40/40 [00:02<00:00, 19.85it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.71it/s]\n",
+                        "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.69it/s]\n"
+                    ]
+                }
+            ],
+            "source": [
+                "tsfresh_builder = TSFreshBuilder(\n",
+                "    num_features=200, memory=15, column_id=\"id\", time_stamp=\"date\", target=\"Occupancy\"\n",
+                ")\n",
+                "\n",
+                "with benchmark(\"tsfresh\"):\n",
+                "    tsfresh_train = tsfresh_builder.fit(dfs_pandas[\"train\"])\n",
+                "\n",
+                "tsfresh_test = tsfresh_builder.transform(dfs_pandas[\"test\"])\n",
+                "\n",
+                "df_tsfresh_train = getml.data.DataFrame.from_pandas(\n",
+                "    tsfresh_train, name=\"tsfresh_train\", roles=data_train.roles\n",
+                ")\n",
+                "df_tsfresh_test = getml.data.DataFrame.from_pandas(\n",
+                "    tsfresh_test, name=\"tsfresh_test\", roles=data_train.roles\n",
+                ")\n",
+                "\n",
+                "df_tsfresh_train.set_role(df_tsfresh_train.roles.unused, getml.data.roles.numerical)\n",
+                "\n",
+                "df_tsfresh_test.set_role(df_tsfresh_test.roles.unused, getml.data.roles.numerical)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 28,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])"
+                        ]
+                    },
+                    "execution_count": 28,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
+                "    tags=[\"predicition\", \"tsfresh\"], predictors=[predictor]\n",
+                ")\n",
+                "\n",
+                "pipe_tsf_pr"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 29,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">label                  </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">message                         </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">WARNING</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    type      label                     message                         \n",
+                            "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
+                        ]
+                    },
+                    "execution_count": 29,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.check(df_tsfresh_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 30,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Checking data model\u001b[33m...\u001b[0m\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
+                            "</pre>\n"
+                        ],
+                        "text/plain": [
+                            "Trained pipeline.\n"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                },
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Time taken: 0:00:01.669099.\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=[],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='SquareLoss',\n",
+                            "         peripheral=[],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['predicition', 'tsfresh'])"
+                        ]
+                    },
+                    "execution_count": 30,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.fit(df_tsfresh_train)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 31,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
+                        "\u001b[?25h"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target   </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">accuracy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    auc</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">cross entropy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:53:00</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9985</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">1.</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.006898</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-09-13 13:53:00</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">tsfresh_test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">Occupancy</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9877</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9979</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.049359</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used        target      accuracy       auc   cross entropy\n",
+                            "0   2024-09-13 13:53:00   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
+                            "1   2024-09-13 13:53:00   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
+                        ]
+                    },
+                    "execution_count": 31,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe_tsf_pr.score(df_tsfresh_test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {
+                "lines_to_next_cell": 2
+            },
+            "source": [
+                "### 3. Comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 32,
+            "metadata": {
+                "lines_to_next_cell": 2
+            },
+            "outputs": [],
+            "source": [
+                "num_features = dict(\n",
+                "    fastprop=289,\n",
+                "    featuretools=103,\n",
+                "    tsfresh=60,\n",
+                ")\n",
+                "\n",
+                "runtime_per_feature = [\n",
+                "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
+                "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
+                "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
+                "]\n",
+                "\n",
+                "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
+                "\n",
+                "normalized_runtime_per_feature = [\n",
+                "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
+                "]\n",
+                "\n",
+                "comparison = pd.DataFrame(\n",
+                "    dict(\n",
+                "        runtime=[\n",
+                "            benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"featuretools\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"],\n",
+                "        ],\n",
+                "        num_features=num_features.values(),\n",
+                "        features_per_second=features_per_second,\n",
+                "        normalized_runtime=[\n",
+                "            1,\n",
+                "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
+                "        ],\n",
+                "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
+                "        accuracy=[pipe_fp_pr.accuracy, pipe_ft_pr.accuracy, pipe_tsf_pr.accuracy],\n",
+                "        auc=[pipe_fp_pr.auc, pipe_ft_pr.auc, pipe_tsf_pr.auc],\n",
+                "        cross_entropy=[\n",
+                "            pipe_fp_pr.cross_entropy,\n",
+                "            pipe_ft_pr.cross_entropy,\n",
+                "            pipe_tsf_pr.cross_entropy,\n",
+                "        ],\n",
+                "    )\n",
+                ")\n",
+                "\n",
+                "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 33,
+            "metadata": {
+                "tags": []
+            },
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>runtime</th>\n",
+                            "      <th>num_features</th>\n",
+                            "      <th>features_per_second</th>\n",
+                            "      <th>normalized_runtime</th>\n",
+                            "      <th>normalized_runtime_per_feature</th>\n",
+                            "      <th>accuracy</th>\n",
+                            "      <th>auc</th>\n",
+                            "      <th>cross_entropy</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>getML: FastProp</th>\n",
+                            "      <td>0 days 00:00:01.825967</td>\n",
+                            "      <td>289</td>\n",
+                            "      <td>158.277936</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>0.988823</td>\n",
+                            "      <td>0.998166</td>\n",
+                            "      <td>0.044213</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>featuretools</th>\n",
+                            "      <td>0 days 00:07:20.110459</td>\n",
+                            "      <td>103</td>\n",
+                            "      <td>0.234032</td>\n",
+                            "      <td>241.028704</td>\n",
+                            "      <td>676.308484</td>\n",
+                            "      <td>0.988455</td>\n",
+                            "      <td>0.997207</td>\n",
+                            "      <td>0.049236</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>tsfresh</th>\n",
+                            "      <td>0 days 00:00:14.295312</td>\n",
+                            "      <td>60</td>\n",
+                            "      <td>4.197184</td>\n",
+                            "      <td>7.828899</td>\n",
+                            "      <td>37.710510</td>\n",
+                            "      <td>0.987718</td>\n",
+                            "      <td>0.997861</td>\n",
+                            "      <td>0.049359</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "                               runtime  num_features  features_per_second  \\\n",
+                            "getML: FastProp 0 days 00:00:01.825967           289           158.277936   \n",
+                            "featuretools    0 days 00:07:20.110459           103             0.234032   \n",
+                            "tsfresh         0 days 00:00:14.295312            60             4.197184   \n",
+                            "\n",
+                            "                 normalized_runtime  normalized_runtime_per_feature  accuracy  \\\n",
+                            "getML: FastProp            1.000000                        1.000000  0.988823   \n",
+                            "featuretools             241.028704                      676.308484  0.988455   \n",
+                            "tsfresh                    7.828899                       37.710510  0.987718   \n",
+                            "\n",
+                            "                      auc  cross_entropy  \n",
+                            "getML: FastProp  0.998166       0.044213  \n",
+                            "featuretools     0.997207       0.049236  \n",
+                            "tsfresh          0.997861       0.049359  "
+                        ]
+                    },
+                    "execution_count": 33,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "comparison"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 34,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# export for further use\n",
+                "comparison.to_csv(\"comparisons/occupancy.csv\")"
+            ]
+        }
+    ],
+    "metadata": {
+        "celltoolbar": "Tags",
+        "jupytext": {
+            "encoding": "# -*- coding: utf-8 -*-",
+            "formats": "ipynb,py:percent"
+        },
+        "kernelspec": {
+            "display_name": "Python 3 (ipykernel)",
+            "language": "python",
+            "name": "python3"
+        },
+        "language_info": {
+            "codemirror_mode": {
+                "name": "ipython",
+                "version": 3
+            },
+            "file_extension": ".py",
+            "mimetype": "text/x-python",
+            "name": "python",
+            "nbconvert_exporter": "python",
+            "pygments_lexer": "ipython3",
+            "version": "3.11.9"
+        },
+        "toc-autonumbering": false,
+        "toc-showcode": false,
+        "toc-showmarkdowntxt": false,
+        "toc-showtags": false,
+        "widgets": {
+            "application/vnd.jupyter.widget-state+json": {
+                "state": {},
+                "version_major": 2,
+                "version_minor": 0
+            }
+        }
     },
-    "tags": [
-     "hide_input"
-    ]
-   },
-   "source": [
-    "# Propositionalization: Occupancy detection\n",
-    "\n",
-    "In this notebbok, we compare getML's FastProp against well-known feature engineering libraries featuretools and tsfresh.\n",
-    "\n",
-    "Summary:\n",
-    "\n",
-    "- Prediction type: __Binary classification__\n",
-    "- Domain: __Energy__\n",
-    "- Prediction target: __Room occupancy__\n",
-    "- Source data: __1 table, 32k rows__\n",
-    "- Population size: __32k__"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "tags": [
-     "remove_cell_on_docs"
-    ]
-   },
-   "source": [
-    "<a target=\"_blank\" href=\"https://colab.research.google.com/github/getml/getml-demo/blob/master/fastprop_benchmark/occupancy_prop.ipynb\">\n",
-    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
-    "</a>"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "## Background\n",
-    "\n",
-    "A common approach to feature engineering is to generate attribute-value representations from relational data by applying a fixed set of aggregations to columns of interest and perform a feature selection on the (possibly large) set of generated features afterwards. In academia, this approach is called _propositionalization._\n",
-    "\n",
-    "getML's [FastProp](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html#fastprop) is an implementation of this propositionalization approach that has been optimized for speed and memory efficiency. In this notebook, we want to demonstrate how – well – fast FastProp is. To this end, we will benchmark FastProp against the popular feature engineering libraries [featuretools](https://www.featuretools.com/) and [tsfresh](https://tsfresh.readthedocs.io/en/latest/). Both of these libraries use propositionalization approaches for feature engineering.\n",
-    "\n",
-    "Our use case here is a public domain data set for predicting room occupancy from sensor data. For further details about the data set refer to [the full notebook](../occupancy.ipynb)."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Analysis"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "1. [Loading data](#1.-Loading-data)\n",
-    "2. [Predictive modeling](#2.-Predictive-modeling)\n",
-    "3. [Comparison](#3.-Comparison)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Let's get started with the analysis and set-up your session:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Note: you may need to restart the kernel to use updated packages.\n"
-     ]
-    }
-   ],
-   "source": [
-    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "getML API version: 1.5.0\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "import os\n",
-    "import sys\n",
-    "\n",
-    "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
-    "from pathlib import Path\n",
-    "\n",
-    "import pandas as pd\n",
-    "import getml\n",
-    "\n",
-    "print(f\"getML API version: {getml.__version__}\\n\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "getML Engine is already running.\n",
-      "\u001b[2K  Loading pipelines... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Connected to project <span style=\"color: #008000; text-decoration-color: #008000\">'occupancy'</span>.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Connected to project \u001b[32m'occupancy'\u001b[0m.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "getml.engine.launch(allow_remote_ips=True, token=\"token\")\n",
-    "getml.engine.set_project(\"occupancy\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# If we are in Colab, we need to fetch the utils folder from the repository\n",
-    "if os.getenv(\"COLAB_RELEASE_TAG\"):\n",
-    "    !curl -L https://api.github.com/repos/getml/getml-demo/tarball/master | tar --wildcards --strip-components=1 -xz '*utils*'"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "parent = Path(os.getcwd()).parent.as_posix()\n",
-    "\n",
-    "if parent not in sys.path:\n",
-    "    sys.path.append(parent)\n",
-    "\n",
-    "from utils import Benchmark, FTTimeSeriesBuilder, TSFreshBuilder"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 1. Loading data\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The data set can be downloaded directly from GitHub. It is conveniently separated into a train, a validation and a testing set. This allows us to directly benchmark our results against the results of the original paper later."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_test, data_train, data_validate = getml.datasets.load_occupancy(roles=True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_all, split = getml.data.split.concat(\n",
-    "    \"data_all\",\n",
-    "    train=data_train,\n",
-    "    validation=data_validate,\n",
-    "    test=data_test,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The train set looks like this:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {
-    "tags": [
-     "hide_input"
-    ]
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th>name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"time_stamp\">               date</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"target\">Occupancy</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">Temperature</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\"> Humidity</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">    Light</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">      CO2</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">HumidityRatio</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\">role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">         time_stamp</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">   target</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">  numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">    numerical</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\">unit</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header time_stamp\">         time stamp</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">           </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">         </th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">             </th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-11 14:48:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.76</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">31</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.1333</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">437</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.3333</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1029</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.6667</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.005021</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-11 14:49:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.79</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">31</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">437</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.3333</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1000</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.005009</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-11 14:50:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.7675</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">31</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.1225</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">434</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1003</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.75</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.005022</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-11 14:51:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.7675</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">31</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.1225</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">439</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1009</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.5</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.005022</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-11 14:51:59</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.79</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">31</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.1333</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">437</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.3333</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1005</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.6667</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.00503</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">...</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>9747</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-18 09:15:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">20</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.815</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">27</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.7175</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">429</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.75</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1505</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.25</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.004213</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>9748</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-18 09:16:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">20</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.865</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">27</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.745</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">423</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.5</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1514</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.5</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.00423</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>9749</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-18 09:16:59</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">20</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.89</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">27</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.745</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">423</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.5</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1521</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.5</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.004237</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>9750</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-18 09:17:59</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">20</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.89</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">28</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.0225</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">418</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.75</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1632</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.004279</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>9751</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"time_stamp\">2015-02-18 09:19:00</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">21</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">28</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >.1</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">409</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">1864</span\n",
-       "              ><span class=\"right\" style=\"width: 4ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "            \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 6ch\"\n",
-       "                >.004321</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    9752 rows x 7 columns<br />\n",
-       "    memory usage: 0.55 MB<br />\n",
-       "    name: population_test<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       "name                  date   Occupancy   Temperature    Humidity       Light         CO2   HumidityRatio\n",
-       "role            time_stamp      target     numerical   numerical   numerical   numerical       numerical\n",
-       "unit            time stamp                                                                              \n",
-       "   0   2015-02-11 14:48:00           1       21.76       31.1333    437.3333   1029.6667        0.005021\n",
-       "   1   2015-02-11 14:49:00           1       21.79       31         437.3333   1000             0.005009\n",
-       "   2   2015-02-11 14:50:00           1       21.7675     31.1225    434        1003.75          0.005022\n",
-       "   3   2015-02-11 14:51:00           1       21.7675     31.1225    439        1009.5           0.005022\n",
-       "   4   2015-02-11 14:51:59           1       21.79       31.1333    437.3333   1005.6667        0.00503 \n",
-       "                       ...         ...       ...         ...         ...         ...           ...      \n",
-       "9747   2015-02-18 09:15:00           1       20.815      27.7175    429.75     1505.25          0.004213\n",
-       "9748   2015-02-18 09:16:00           1       20.865      27.745     423.5      1514.5           0.00423 \n",
-       "9749   2015-02-18 09:16:59           1       20.89       27.745     423.5      1521.5           0.004237\n",
-       "9750   2015-02-18 09:17:59           1       20.89       28.0225    418.75     1632             0.004279\n",
-       "9751   2015-02-18 09:19:00           1       21          28.1       409        1864             0.004321\n",
-       "\n",
-       "\n",
-       "9752 rows x 7 columns\n",
-       "memory usage: 0.55 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_train"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### 2. Predictive modeling\n",
-    "\n",
-    "We loaded the data, defined the roles, units and the abstract data model. Next, we create a getML pipeline for relational learning."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.1 Propositionalization with getML's FastProp"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We use all possible aggregations. Because tsfresh and featuretools are single-threaded, we limit our FastProp algorithm to one thread as well, to ensure a fair comparison."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
-       "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
-       "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">data_all</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"40.0\" x=\"324.0\" font-size=\"7pt\" >date &lt;= date</tspan><tspan y=\"50.0\" x=\"324.0\" font-size=\"7pt\" >Memory: 15.0 minutes</tspan></text></svg></div>\n",
-       "            </div>\n",
-       "\n",
-       "            <div style='margin-top: 15px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
-       "            <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">data frames</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">staging table              </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_all</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DATA_ALL__STAGING_TABLE_2</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "            </div>\n",
-       "            \n",
-       "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
-       "<div style='margin-top: 15px;'>\n",
-       "<div style='float: left; margin-right: 50px;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">subset    </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name    </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">rows   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_all</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_all</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">validation</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_all</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">unknown</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "<div style='float: left;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name    </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> rows</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">data_all</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">20560</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DataFrame</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "data model\n",
-       "\n",
-       "  population:\n",
-       "    columns:\n",
-       "    - Temperature: numerical\n",
-       "    - Humidity: numerical\n",
-       "    - Light: numerical\n",
-       "    - CO2: numerical\n",
-       "    - HumidityRatio: numerical\n",
-       "    - ...\n",
-       "\n",
-       "    joins:\n",
-       "    - right: 'data_all'\n",
-       "      time_stamps: (population.date, data_all.date)\n",
-       "      relationship: 'many-to-many'\n",
-       "      memory: 900.0\n",
-       "      horizon: 0.0\n",
-       "      lagged_targets: False\n",
-       "\n",
-       "  data_all:\n",
-       "    columns:\n",
-       "    - Temperature: numerical\n",
-       "    - Humidity: numerical\n",
-       "    - Light: numerical\n",
-       "    - CO2: numerical\n",
-       "    - HumidityRatio: numerical\n",
-       "    - ...\n",
-       "\n",
-       "\n",
-       "container\n",
-       "\n",
-       "  population\n",
-       "      subset       name       rows      type\n",
-       "  0   test         data_all   unknown   View\n",
-       "  1   train        data_all   unknown   View\n",
-       "  2   validation   data_all   unknown   View\n",
-       "\n",
-       "  peripheral\n",
-       "      name        rows   type     \n",
-       "  0   data_all   20560   DataFrame"
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# Our forecast horizon is 0.\n",
-    "# We do not predict the future, instead we infer\n",
-    "# the present state from current and past sensor data.\n",
-    "horizon = 0.0\n",
-    "\n",
-    "# We do not allow the time series features\n",
-    "# to use target values from the past.\n",
-    "# (Otherwise, we would need the horizon to\n",
-    "# be greater than 0.0).\n",
-    "allow_lagged_targets = False\n",
-    "\n",
-    "# We want our time series features to only use\n",
-    "# data from the last 15 minutes\n",
-    "memory = getml.data.time.minutes(15)\n",
-    "\n",
-    "time_series = getml.data.TimeSeries(\n",
-    "    population=data_all,\n",
-    "    split=split,\n",
-    "    time_stamps=\"date\",\n",
-    "    horizon=horizon,\n",
-    "    memory=memory,\n",
-    "    lagged_targets=allow_lagged_targets,\n",
-    ")\n",
-    "\n",
-    "time_series"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "feature_learner = getml.feature_learning.FastProp(\n",
-    "    loss_function=getml.feature_learning.loss_functions.CrossEntropyLoss,\n",
-    "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
-    "    num_threads=1,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Next, we create the pipeline. In contrast to our usual approach, we create _two pipelines_ in\n",
-    "this notebook. One for feature learning (suffix `_fl`) and one for predicition (suffix `_pr`).\n",
-    "This allows for a fair comparison of runtimes."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "pipe_fp_fl = getml.pipeline.Pipeline(\n",
-    "    feature_learners=[feature_learner],\n",
-    "    data_model=time_series.data_model,\n",
-    "    tags=[\"feature learning\", \"fastprop\"],\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pipe_fp_fl.check(time_series.train)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The wrappers around featuretools and tsfresh fit on the training set and then return the training features. We therefore measure the time it takes getML's FastProp algorithm to fit on the training set and create the training features."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "benchmark = Benchmark()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Trying 331 features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:01.047819.\n",
-      "\n",
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"fastprop\"):\n",
-    "    pipe_fp_fl.fit(time_series.train)\n",
-    "    fastprop_train = pipe_fp_fl.transform(time_series.train, df_name=\"fastprop_train\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  FastProp: Building features... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    }
-   ],
-   "source": [
-    "fastprop_test = pipe_fp_fl.transform(time_series.test, df_name=\"fastprop_test\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now we create a dedicated prediction pipeline and provide the fast prop features\n",
-    "(contrained in `fastprop_train` and `fastprop_test`.)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "predictor = getml.predictors.XGBoostClassifier()\n",
-    "\n",
-    "pipe_fp_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"prediction\", \"fastprop\"], predictors=[predictor]\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">OK.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "OK.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:04.853148.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'fastprop'])"
-      ]
-     },
-     "execution_count": 17,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.check(fastprop_train)\n",
-    "\n",
-    "pipe_fp_pr.fit(fastprop_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used      </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">accuracy</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    auc</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">cross entropy</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:20:28</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9995</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">1.</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.004464</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:20:29</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">fastprop_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9888</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9982</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.044213</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used         target      accuracy       auc   cross entropy\n",
-       "0   2024-09-12 16:20:28   fastprop_train   Occupancy     0.9995    1.            0.004464\n",
-       "1   2024-09-12 16:20:29   fastprop_test    Occupancy     0.9888    0.9982        0.044213"
-      ]
-     },
-     "execution_count": 18,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_fp_pr.score(fastprop_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.2 Propositionalization with featuretools"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_train = time_series.train.population.to_df(\"train\")\n",
-    "data_test = time_series.test.population.to_df(\"test\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "dfs_pandas = {}\n",
-    "\n",
-    "for df in getml.project.data_frames:\n",
-    "    dfs_pandas[df.name] = df.to_pandas()\n",
-    "    dfs_pandas[df.name][\"id\"] = 1"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "ft_builder = FTTimeSeriesBuilder(\n",
-    "    num_features=200,\n",
-    "    horizon=pd.Timedelta(minutes=0),\n",
-    "    memory=pd.Timedelta(minutes=15),\n",
-    "    column_id=\"id\",\n",
-    "    time_stamp=\"date\",\n",
-    "    target=\"Occupancy\",\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The `FTTimeSeriesBuilder` provides a `fit` method that is designed to be equivilant to\n",
-    "to the `fit` method of the predictorless getML pipeline above."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "featuretools: Trying features...\n",
-      "Selecting the best out of 262 features...\n",
-      "Time taken: 0h:7m:28.128247\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "with benchmark(\"featuretools\"):\n",
-    "    featuretools_train = ft_builder.fit(dfs_pandas[\"train\"])\n",
-    "\n",
-    "featuretools_test = ft_builder.transform(dfs_pandas[\"test\"])\n",
-    "\n",
-    "df_featuretools_train = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_train, name=\"featuretools_train\", roles=data_train.roles\n",
-    ")\n",
-    "df_featuretools_test = getml.data.DataFrame.from_pandas(\n",
-    "    featuretools_test, name=\"featuretools_test\", roles=data_train.roles\n",
-    ")\n",
-    "\n",
-    "df_featuretools_train.set_role(\n",
-    "    df_featuretools_train.roles.unused, getml.data.roles.numerical\n",
-    ")\n",
-    "\n",
-    "df_featuretools_test.set_role(\n",
-    "    df_featuretools_test.roles.unused, getml.data.roles.numerical\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])"
-      ]
-     },
-     "execution_count": 23,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "predictor = getml.predictors.XGBoostClassifier()\n",
-    "\n",
-    "pipe_ft_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"prediction\", \"featuretools\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_ft_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">label                  </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">message                         </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">WARNING</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    type      label                     message                         \n",
-       "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
-      ]
-     },
-     "execution_count": 24,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.check(df_featuretools_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:03.326992.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['prediction', 'featuretools'])"
-      ]
-     },
-     "execution_count": 25,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.fit(df_featuretools_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">accuracy</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    auc</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">cross entropy</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:34:15</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9995</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">1.</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.005065</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:34:16</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">featuretools_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9885</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9972</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.049236</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used             target      accuracy       auc   cross entropy\n",
-       "0   2024-09-12 16:34:15   featuretools_train   Occupancy     0.9995    1.            0.005065\n",
-       "1   2024-09-12 16:34:16   featuretools_test    Occupancy     0.9885    0.9972        0.049236"
-      ]
-     },
-     "execution_count": 26,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_ft_pr.score(df_featuretools_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "#### 2.3 Propositionalization with tsfresh"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {
-    "lines_to_next_cell": 2
-   },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:03<00:00, 12.60it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:06<00:00,  6.46it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:05<00:00,  7.32it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Selecting the best out of 65 features...\n",
-      "Time taken: 0h:0m:17.070296\n",
-      "\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 19.76it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.89it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.92it/s]\n"
-     ]
-    }
-   ],
-   "source": [
-    "tsfresh_builder = TSFreshBuilder(\n",
-    "    num_features=200, memory=15, column_id=\"id\", time_stamp=\"date\", target=\"Occupancy\"\n",
-    ")\n",
-    "\n",
-    "with benchmark(\"tsfresh\"):\n",
-    "    tsfresh_train = tsfresh_builder.fit(dfs_pandas[\"train\"])\n",
-    "\n",
-    "tsfresh_test = tsfresh_builder.transform(dfs_pandas[\"test\"])\n",
-    "\n",
-    "df_tsfresh_train = getml.data.DataFrame.from_pandas(\n",
-    "    tsfresh_train, name=\"tsfresh_train\", roles=data_train.roles\n",
-    ")\n",
-    "df_tsfresh_test = getml.data.DataFrame.from_pandas(\n",
-    "    tsfresh_test, name=\"tsfresh_test\", roles=data_train.roles\n",
-    ")\n",
-    "\n",
-    "df_tsfresh_train.set_role(df_tsfresh_train.roles.unused, getml.data.roles.numerical)\n",
-    "\n",
-    "df_tsfresh_test.set_role(df_tsfresh_test.roles.unused, getml.data.roles.numerical)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 28,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])"
-      ]
-     },
-     "execution_count": 28,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr = getml.pipeline.Pipeline(\n",
-    "    tags=[\"predicition\", \"tsfresh\"], predictors=[predictor]\n",
-    ")\n",
-    "\n",
-    "pipe_tsf_pr"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 29,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Checking... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">label                  </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">message                         </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">WARNING</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">COLUMN SHOULD BE UNUSED</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;id&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    type      label                     message                         \n",
-       "0   WARNING   COLUMN SHOULD BE UNUSED   All non-NULL entries in column '..."
-      ]
-     },
-     "execution_count": 29,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.check(df_tsfresh_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Checking data model<span style=\"color: #808000; text-decoration-color: #808000\">...</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Checking data model\u001b[33m...\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">The pipeline check generated <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0</span> issues labeled INFO and <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span> issues labeled WARNING.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "The pipeline check generated \u001b[1;36m0\u001b[0m issues labeled INFO and \u001b[1;36m1\u001b[0m issues labeled WARNING.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">To see the issues in full, run <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">.check</span><span style=\"font-weight: bold\">()</span> on the pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "To see the issues in full, run \u001b[1;35m.check\u001b[0m\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m on the pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:01\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Trained pipeline.\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Trained pipeline.\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Time taken: 0:00:01.784330.\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=[],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='SquareLoss',\n",
-       "         peripheral=[],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['predicition', 'tsfresh'])"
-      ]
-     },
-     "execution_count": 30,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.fit(df_tsfresh_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 31,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[?25h"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used     </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target   </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">accuracy</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    auc</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">cross entropy</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:34:49</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9985</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">1.</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.006898</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-09-12 16:34:49</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">tsfresh_test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">Occupancy</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9877</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9979</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.049359</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used        target      accuracy       auc   cross entropy\n",
-       "0   2024-09-12 16:34:49   tsfresh_train   Occupancy     0.9985    1.            0.006898\n",
-       "1   2024-09-12 16:34:49   tsfresh_test    Occupancy     0.9877    0.9979        0.049359"
-      ]
-     },
-     "execution_count": 31,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe_tsf_pr.score(df_tsfresh_test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "lines_to_next_cell": 2
-   },
-   "source": [
-    "### 3. Comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {
-    "lines_to_next_cell": 2
-   },
-   "outputs": [],
-   "source": [
-    "num_features = dict(\n",
-    "    fastprop=289,\n",
-    "    featuretools=103,\n",
-    "    tsfresh=60,\n",
-    ")\n",
-    "\n",
-    "runtime_per_feature = [\n",
-    "    benchmark.runtimes[\"fastprop\"] / num_features[\"fastprop\"],\n",
-    "    benchmark.runtimes[\"featuretools\"] / num_features[\"featuretools\"],\n",
-    "    benchmark.runtimes[\"tsfresh\"] / num_features[\"tsfresh\"],\n",
-    "]\n",
-    "\n",
-    "features_per_second = [1.0 / r.total_seconds() for r in runtime_per_feature]\n",
-    "\n",
-    "normalized_runtime_per_feature = [\n",
-    "    r / runtime_per_feature[0] for r in runtime_per_feature\n",
-    "]\n",
-    "\n",
-    "comparison = pd.DataFrame(\n",
-    "    dict(\n",
-    "        runtime=[\n",
-    "            benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"featuretools\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"],\n",
-    "        ],\n",
-    "        num_features=num_features.values(),\n",
-    "        features_per_second=features_per_second,\n",
-    "        normalized_runtime=[\n",
-    "            1,\n",
-    "            benchmark.runtimes[\"featuretools\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "            benchmark.runtimes[\"tsfresh\"] / benchmark.runtimes[\"fastprop\"],\n",
-    "        ],\n",
-    "        normalized_runtime_per_feature=normalized_runtime_per_feature,\n",
-    "        accuracy=[pipe_fp_pr.accuracy, pipe_ft_pr.accuracy, pipe_tsf_pr.accuracy],\n",
-    "        auc=[pipe_fp_pr.auc, pipe_ft_pr.auc, pipe_tsf_pr.auc],\n",
-    "        cross_entropy=[\n",
-    "            pipe_fp_pr.cross_entropy,\n",
-    "            pipe_ft_pr.cross_entropy,\n",
-    "            pipe_tsf_pr.cross_entropy,\n",
-    "        ],\n",
-    "    )\n",
-    ")\n",
-    "\n",
-    "comparison.index = [\"getML: FastProp\", \"featuretools\", \"tsfresh\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>runtime</th>\n",
-       "      <th>num_features</th>\n",
-       "      <th>features_per_second</th>\n",
-       "      <th>normalized_runtime</th>\n",
-       "      <th>normalized_runtime_per_feature</th>\n",
-       "      <th>accuracy</th>\n",
-       "      <th>auc</th>\n",
-       "      <th>cross_entropy</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:01.854957</td>\n",
-       "      <td>289</td>\n",
-       "      <td>155.787506</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>0.988823</td>\n",
-       "      <td>0.998166</td>\n",
-       "      <td>0.044213</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>featuretools</th>\n",
-       "      <td>0 days 00:07:28.129456</td>\n",
-       "      <td>103</td>\n",
-       "      <td>0.229844</td>\n",
-       "      <td>241.584822</td>\n",
-       "      <td>677.795763</td>\n",
-       "      <td>0.988455</td>\n",
-       "      <td>0.997207</td>\n",
-       "      <td>0.049236</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:17.070465</td>\n",
-       "      <td>60</td>\n",
-       "      <td>3.514840</td>\n",
-       "      <td>9.202620</td>\n",
-       "      <td>44.322792</td>\n",
-       "      <td>0.987718</td>\n",
-       "      <td>0.997861</td>\n",
-       "      <td>0.049359</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:01.854957           289           155.787506   \n",
-       "featuretools    0 days 00:07:28.129456           103             0.229844   \n",
-       "tsfresh         0 days 00:00:17.070465            60             3.514840   \n",
-       "\n",
-       "                 normalized_runtime  normalized_runtime_per_feature  accuracy  \\\n",
-       "getML: FastProp            1.000000                        1.000000  0.988823   \n",
-       "featuretools             241.584822                      677.795763  0.988455   \n",
-       "tsfresh                    9.202620                       44.322792  0.987718   \n",
-       "\n",
-       "                      auc  cross_entropy  \n",
-       "getML: FastProp  0.998166       0.044213  \n",
-       "featuretools     0.997207       0.049236  \n",
-       "tsfresh          0.997861       0.049359  "
-      ]
-     },
-     "execution_count": 33,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "comparison"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 34,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# export for further use\n",
-    "comparison.to_csv(\"comparisons/occupancy.csv\")"
-   ]
-  }
- ],
- "metadata": {
-  "celltoolbar": "Tags",
-  "jupytext": {
-   "encoding": "# -*- coding: utf-8 -*-",
-   "formats": "ipynb,py:percent"
-  },
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.9"
-  },
-  "toc-autonumbering": false,
-  "toc-showcode": false,
-  "toc-showmarkdowntxt": false,
-  "toc-showtags": false,
-  "widgets": {
-   "application/vnd.jupyter.widget-state+json": {
-    "state": {},
-    "version_major": 2,
-    "version_minor": 0
-   }
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
+    "nbformat": 4,
+    "nbformat_minor": 4
 }
diff --git a/fastprop_benchmark/robot_prop.ipynb b/fastprop_benchmark/robot_prop.ipynb
index 3f29ccb..4536895 100644
--- a/fastprop_benchmark/robot_prop.ipynb
+++ b/fastprop_benchmark/robot_prop.ipynb
@@ -82,7 +82,7 @@
     }
    ],
    "source": [
-    "%pip install -q \"getml==1.5.0\" \"featuretools==1.28.0\" \"tsfresh==0.20.2\""
+    "%pip install -q \"getml==1.5.0\" \"featuretools==1.31.0\" \"tsfresh==0.20.3\""
    ]
   },
   {
@@ -30233,7 +30233,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:00.043565.\n",
+      "Time taken: 0:00:00.032192.\n",
       "\n",
       "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
       "\u001b[2K  Preprocessing... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
@@ -30403,7 +30403,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_122&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_126&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30420,7 +30420,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_123&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_127&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30437,7 +30437,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_127&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_128&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30454,7 +30454,7 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_129&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
+       "              <td class=\"str\">All non-NULL entries in column &#x27;feature_1_132&#x27; in POPULATION__STAGING_TABLE_1 are equal to each other. You should consider setting its role to unused_float or using it for comparison only (you can do the latter by setting a unit that contains &#x27;comparison only&#x27;).</td>\n",
        "            \n",
        "          \n",
        "      </tr>\n",
@@ -30558,7 +30558,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.089082.\n",
+      "Time taken: 0:00:04.112529.\n",
       "\n"
      ]
     },
@@ -30681,7 +30681,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:26:03</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:16:39</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30710,7 +30710,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:26:03</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:16:39</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -30740,8 +30740,8 @@
       ],
       "text/plain": [
        "    date time             set used         target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:26:03   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
-       "1   2024-09-12 17:26:03   fastprop_test    f_x       0.5516    0.7237     0.9951"
+       "0   2024-09-13 14:16:39   fastprop_train   f_x       0.4383    0.5764     0.9963\n",
+       "1   2024-09-13 14:16:39   fastprop_test    f_x       0.5516    0.7237     0.9951"
       ]
      },
      "execution_count": 21,
@@ -31080,7 +31080,7 @@
      "text": [
       "featuretools: Trying features...\n",
       "Selecting the best out of 442 features...\n",
-      "Time taken: 0h:14m:36.192123\n",
+      "Time taken: 0h:14m:21.609569\n",
       "\n"
      ]
     }
@@ -31182,10 +31182,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
        "      <td>-11.0300</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:00</td>\n",
@@ -31206,10 +31206,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
        "      <td>-10.8480</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:01</td>\n",
@@ -31230,10 +31230,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
        "      <td>-10.6660</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:02</td>\n",
@@ -31254,10 +31254,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
        "      <td>-10.5070</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:03</td>\n",
@@ -31278,10 +31278,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
-       "      <td>1.726162e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
+       "      <td>1.726237e+09</td>\n",
        "      <td>-10.4130</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 00:00:04</td>\n",
@@ -31326,10 +31326,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
        "      <td>-9.7673</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:55</td>\n",
@@ -31350,10 +31350,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
        "      <td>-9.9200</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:56</td>\n",
@@ -31374,10 +31374,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
        "      <td>-9.7743</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:57</td>\n",
@@ -31398,10 +31398,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
        "      <td>-8.6109</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:58</td>\n",
@@ -31422,10 +31422,10 @@
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>True</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
-       "      <td>1.726151e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
+       "      <td>1.726227e+09</td>\n",
        "      <td>-8.4345</td>\n",
        "      <td>1</td>\n",
        "      <td>1970-01-01 02:54:59</td>\n",
@@ -31550,59 +31550,59 @@
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 4)  \\\n",
        "_featuretools_index                                          \n",
-       "0                                             1.726162e+09   \n",
-       "1                                             1.726162e+09   \n",
-       "2                                             1.726162e+09   \n",
-       "3                                             1.726162e+09   \n",
-       "4                                             1.726162e+09   \n",
+       "0                                             1.726237e+09   \n",
+       "1                                             1.726237e+09   \n",
+       "2                                             1.726237e+09   \n",
+       "3                                             1.726237e+09   \n",
+       "4                                             1.726237e+09   \n",
        "...                                                    ...   \n",
-       "10495                                         1.726151e+09   \n",
-       "10496                                         1.726151e+09   \n",
-       "10497                                         1.726151e+09   \n",
-       "10498                                         1.726151e+09   \n",
-       "10499                                         1.726151e+09   \n",
+       "10495                                         1.726227e+09   \n",
+       "10496                                         1.726227e+09   \n",
+       "10497                                         1.726227e+09   \n",
+       "10498                                         1.726227e+09   \n",
+       "10499                                         1.726227e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 30)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.726162e+09   \n",
-       "1                                              1.726162e+09   \n",
-       "2                                              1.726162e+09   \n",
-       "3                                              1.726162e+09   \n",
-       "4                                              1.726162e+09   \n",
+       "0                                              1.726237e+09   \n",
+       "1                                              1.726237e+09   \n",
+       "2                                              1.726237e+09   \n",
+       "3                                              1.726237e+09   \n",
+       "4                                              1.726237e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.726151e+09   \n",
-       "10496                                          1.726151e+09   \n",
-       "10497                                          1.726151e+09   \n",
-       "10498                                          1.726151e+09   \n",
-       "10499                                          1.726151e+09   \n",
+       "10495                                          1.726227e+09   \n",
+       "10496                                          1.726227e+09   \n",
+       "10497                                          1.726227e+09   \n",
+       "10498                                          1.726227e+09   \n",
+       "10499                                          1.726227e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 77)  \\\n",
        "_featuretools_index                                           \n",
-       "0                                              1.726162e+09   \n",
-       "1                                              1.726162e+09   \n",
-       "2                                              1.726162e+09   \n",
-       "3                                              1.726162e+09   \n",
-       "4                                              1.726162e+09   \n",
+       "0                                              1.726237e+09   \n",
+       "1                                              1.726237e+09   \n",
+       "2                                              1.726237e+09   \n",
+       "3                                              1.726237e+09   \n",
+       "4                                              1.726237e+09   \n",
        "...                                                     ...   \n",
-       "10495                                          1.726151e+09   \n",
-       "10496                                          1.726151e+09   \n",
-       "10497                                          1.726151e+09   \n",
-       "10498                                          1.726151e+09   \n",
-       "10499                                          1.726151e+09   \n",
+       "10495                                          1.726227e+09   \n",
+       "10496                                          1.726227e+09   \n",
+       "10497                                          1.726227e+09   \n",
+       "10498                                          1.726227e+09   \n",
+       "10499                                          1.726227e+09   \n",
        "\n",
        "                     TIME_SINCE_LAST_MAX(peripheral.ds, 7)      f_x  id  \\\n",
        "_featuretools_index                                                       \n",
-       "0                                             1.726162e+09 -11.0300   1   \n",
-       "1                                             1.726162e+09 -10.8480   1   \n",
-       "2                                             1.726162e+09 -10.6660   1   \n",
-       "3                                             1.726162e+09 -10.5070   1   \n",
-       "4                                             1.726162e+09 -10.4130   1   \n",
+       "0                                             1.726237e+09 -11.0300   1   \n",
+       "1                                             1.726237e+09 -10.8480   1   \n",
+       "2                                             1.726237e+09 -10.6660   1   \n",
+       "3                                             1.726237e+09 -10.5070   1   \n",
+       "4                                             1.726237e+09 -10.4130   1   \n",
        "...                                                    ...      ...  ..   \n",
-       "10495                                         1.726151e+09  -9.7673   1   \n",
-       "10496                                         1.726151e+09  -9.9200   1   \n",
-       "10497                                         1.726151e+09  -9.7743   1   \n",
-       "10498                                         1.726151e+09  -8.6109   1   \n",
-       "10499                                         1.726151e+09  -8.4345   1   \n",
+       "10495                                         1.726227e+09  -9.7673   1   \n",
+       "10496                                         1.726227e+09  -9.9200   1   \n",
+       "10497                                         1.726227e+09  -9.7743   1   \n",
+       "10498                                         1.726227e+09  -8.6109   1   \n",
+       "10499                                         1.726227e+09  -8.4345   1   \n",
        "\n",
        "                                     ds  \n",
        "_featuretools_index                      \n",
@@ -31779,7 +31779,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.582992.\n",
+      "Time taken: 0:00:04.476476.\n",
       "\n"
      ]
     },
@@ -31902,7 +31902,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:47:07</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:37:28</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31914,11 +31914,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.4394</td>\n",
+       "              <td class=\"float\">0.4396</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.5831</td>\n",
+       "              <td class=\"float\">0.584</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31931,7 +31931,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:47:07</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:37:28</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31943,11 +31943,11 @@
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.571</td>\n",
+       "              <td class=\"float\">0.5828</td>\n",
        "            \n",
        "          \n",
        "            \n",
-       "              <td class=\"float\">0.7484</td>\n",
+       "              <td class=\"float\">0.7595</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -31961,8 +31961,8 @@
       ],
       "text/plain": [
        "    date time             set used             target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:47:07   featuretools_train   f_x       0.4394    0.5831     0.9962\n",
-       "1   2024-09-12 17:47:07   featuretools_test    f_x       0.571     0.7484     0.9948"
+       "0   2024-09-13 14:37:28   featuretools_train   f_x       0.4396    0.584      0.9962\n",
+       "1   2024-09-13 14:37:28   featuretools_test    f_x       0.5828    0.7595     0.9948"
       ]
      },
      "execution_count": 32,
@@ -32005,9 +32005,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:04<00:00,  9.82it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:12<00:00,  3.14it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:11<00:00,  3.60it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:02<00:00, 13.60it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:09<00:00,  4.09it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:11<00:00,  3.64it/s]\n"
      ]
     },
     {
@@ -32015,7 +32015,7 @@
      "output_type": "stream",
      "text": [
       "Selecting the best out of 130 features...\n",
-      "Time taken: 0h:0m:31.641678\n",
+      "Time taken: 0h:0m:26.638195\n",
       "\n"
      ]
     },
@@ -32023,9 +32023,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 28.10it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.14it/s]\n",
-      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  8.33it/s]\n"
+      "Rolling: 100%|██████████| 40/40 [00:01<00:00, 37.19it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.62it/s]\n",
+      "Feature Extraction: 100%|██████████| 40/40 [00:04<00:00,  9.93it/s]\n"
      ]
     }
    ],
@@ -32202,7 +32202,7 @@
      "output_type": "stream",
      "text": [
       "\u001b[2K  Staging... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:00\n",
-      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:04\n",
+      "\u001b[2K  XGBoost: Training as predictor... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% • 00:03\n",
       "\u001b[?25h"
      ]
     },
@@ -32223,7 +32223,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Time taken: 0:00:04.062273.\n",
+      "Time taken: 0:00:03.942466.\n",
       "\n"
      ]
     },
@@ -32346,7 +32346,7 @@
        "        <th>0</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:47:59</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:38:12</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32375,7 +32375,7 @@
        "        <th>1</th>\n",
        "          \n",
        "            \n",
-       "              <td class=\"datetime\">2024-09-12 17:47:59</td>\n",
+       "              <td class=\"datetime\">2024-09-13 14:38:12</td>\n",
        "            \n",
        "          \n",
        "            \n",
@@ -32405,8 +32405,8 @@
       ],
       "text/plain": [
        "    date time             set used        target       mae      rmse   rsquared\n",
-       "0   2024-09-12 17:47:59   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
-       "1   2024-09-12 17:47:59   tsfresh_test    f_x       0.5986    0.7906     0.9938"
+       "0   2024-09-13 14:38:12   tsfresh_train   f_x       0.4916    0.6636     0.9951\n",
+       "1   2024-09-13 14:38:12   tsfresh_test    f_x       0.5986    0.7906     0.9938"
       ]
      },
      "execution_count": 40,
@@ -32512,33 +32512,33 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>getML: FastProp</th>\n",
-       "      <td>0 days 00:00:00.434939</td>\n",
+       "      <td>0 days 00:00:00.398347</td>\n",
        "      <td>134</td>\n",
-       "      <td>308.071473</td>\n",
+       "      <td>336.360579</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>1.000000</td>\n",
        "      <td>0.995058</td>\n",
-       "      <td>0.723674</td>\n",
-       "      <td>0.551593</td>\n",
+       "      <td>0.723716</td>\n",
+       "      <td>0.551608</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>featuretools</th>\n",
-       "      <td>0 days 00:14:36.193549</td>\n",
+       "      <td>0 days 00:14:21.611109</td>\n",
        "      <td>158</td>\n",
-       "      <td>0.180325</td>\n",
-       "      <td>2014.520540</td>\n",
-       "      <td>1708.419285</td>\n",
-       "      <td>0.994832</td>\n",
-       "      <td>0.748423</td>\n",
-       "      <td>0.571029</td>\n",
+       "      <td>0.183377</td>\n",
+       "      <td>2162.966230</td>\n",
+       "      <td>1834.253280</td>\n",
+       "      <td>0.994784</td>\n",
+       "      <td>0.759460</td>\n",
+       "      <td>0.582831</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tsfresh</th>\n",
-       "      <td>0 days 00:00:31.641862</td>\n",
+       "      <td>0 days 00:00:26.638362</td>\n",
        "      <td>120</td>\n",
-       "      <td>3.792447</td>\n",
-       "      <td>72.750114</td>\n",
-       "      <td>81.232902</td>\n",
+       "      <td>4.504789</td>\n",
+       "      <td>66.872255</td>\n",
+       "      <td>74.667339</td>\n",
        "      <td>0.993836</td>\n",
        "      <td>0.790602</td>\n",
        "      <td>0.598600</td>\n",
@@ -32549,18 +32549,18 @@
       ],
       "text/plain": [
        "                               runtime  num_features  features_per_second  \\\n",
-       "getML: FastProp 0 days 00:00:00.434939           134           308.071473   \n",
-       "featuretools    0 days 00:14:36.193549           158             0.180325   \n",
-       "tsfresh         0 days 00:00:31.641862           120             3.792447   \n",
+       "getML: FastProp 0 days 00:00:00.398347           134           336.360579   \n",
+       "featuretools    0 days 00:14:21.611109           158             0.183377   \n",
+       "tsfresh         0 days 00:00:26.638362           120             4.504789   \n",
        "\n",
        "                 normalized_runtime  normalized_runtime_per_feature  rsquared  \\\n",
        "getML: FastProp            1.000000                        1.000000  0.995058   \n",
-       "featuretools            2014.520540                     1708.419285  0.994832   \n",
-       "tsfresh                   72.750114                       81.232902  0.993836   \n",
+       "featuretools            2162.966230                     1834.253280  0.994784   \n",
+       "tsfresh                   66.872255                       74.667339  0.993836   \n",
        "\n",
        "                     rmse       mae  \n",
-       "getML: FastProp  0.723674  0.551593  \n",
-       "featuretools     0.748423  0.571029  \n",
+       "getML: FastProp  0.723716  0.551608  \n",
+       "featuretools     0.759460  0.582831  \n",
        "tsfresh          0.790602  0.598600  "
       ]
      },
diff --git a/kaggle_notebooks/epilepsy_recognition.ipynb b/kaggle_notebooks/epilepsy_recognition.ipynb
index 338496c..62d0f9f 100644
--- a/kaggle_notebooks/epilepsy_recognition.ipynb
+++ b/kaggle_notebooks/epilepsy_recognition.ipynb
@@ -1,4562 +1,4562 @@
 {
- "cells": [
-  {
-   "attachments": {
-    "Linkedin_Optimized_Cover.png": {
-     "image/png": "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"
-    }
-   },
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# \n",
-    "![Using getML on EEG data to classify epileptic seizures](attachment:Linkedin_Optimized_Cover.png)"
-   ]
-  },
-  {
-   "attachments": {
-    "Examples2.png": {
-     "image/png": "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"
-    }
-   },
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Introduction\n",
-    "\n",
-    "### A Primer on Time Series\n",
-    "\n",
-    "Wikipedia defines a time series as follows:\n",
-    "\n",
-    "In mathematics, a time series is a series of data points indexed (or listed or graphed) in time order. Most commonly, a time series is a sequence taken at successive equally spaced points in time.\n",
-    "\n",
-    "The first sentence defines a fundamental feature of time series: your data can be ordered in time. This does not necessitate the existence of a specific time stamp. It only means that data points should be in chronological order.\n",
-    "\n",
-    "Consider the following three examples. All of them are time series. The first example shows force measurements on a robot's arm. We do not know the specific time and date when these measurements were taken. But what we do know is that the data points are indexed according to chronological order. The second example shows hourly traffic volume on interstate 94 (a highway in the United States of America) in the first week of April 2016. This data has an exact date and time when the measurements were taken. The third example is a bio-signal, a recording of an EEG signal of human brain activity. For this kind of data, a specific time stamp can be useful, for example sleep data over longer time periods such as months, but commonly is not necessary for short term bio-signals such as EEG readings or movement data.\n",
-    "\n",
-    "![Examples plots](attachment:Examples2.png)\n",
-    "\n",
-    "The second sentence in Wikipedia's definition says that points in time are most commonly equally spaced. This in fact holds true for most time series data. We know for a fact that traffic volume was measured once every hour. Although we do not have a time stamp for the robot arm data, we can assume that the data recording sensor measured in equally spaced time intervals. \n",
-    "\n",
-    "As a matter of fact, even these equidistant time intervals might be an assumption in data sets like the above examples, as all sensors physically record data points with some form of timely variance due to multiple reasons such as sensor or data saving lags and so on. Additionally, there might be an inherent variance; for example a person never sleeps at exactly the same time every day. However, in many real-world data sets you will find that the assumption of equally spaced data points in time is violated and you will have to deal with that."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Time Series Prediction Tasks\n",
-    "\n",
-    "There are several kinds of time series problems, but they can be roughly subdivided into two kinds: *Forecasts* and *Nowcasts*.\n",
-    "\n",
-    "- A *forecast* tries to predict the future. For instance, when dealing with the Interstate 94 data, you will most likely want to predict the traffic volume in one hour from now or one week from now. This time period is often called the *forecast horizon*.\n",
-    "\n",
-    "- A *nowcast* tries to predict another time series or class-label based on some time series. For instance, when confronted with the robot data, you might want to predict the force vector based on other sensor data. This is useful when other sensor data are easier to measure than the force vector. On the other hand, a very common task for bio-signals is classifying some state based on these signals, such as sleep quality or epileptic seizure occurance.\n",
-    "\n",
-    "A very interesting question is also whether you are allowed to use lagged targets. For instance, if you want to predict the traffic volume for the next hour, are you allowed to use the traffic volume for the last three hours? In this case, probably yes, but if you want to do a nowcast on the force vector of the robot's arm, probably no, because the entire idea is that measuring the force vector is difficult and therefore we cannot assume that we have the force vector data of the recent past when we are making a prediction. This question is also very relevant for classification tasks, as you face the question of how far in the past data is relevant to classifying a specific time point or interval. For example, in sleep data classification, events that occurred several days, weeks, or even months in the past might have an influence, while an epileptic seizure might be detectable in the short term."
-   ]
-  },
-  {
-   "attachments": {
-    "ManualFeatureEngineering2.png": {
-     "image/png": "iVBORw0KGgoAAAANSUhEUgAABZMAAAIeCAYAAADOA5F4AAAACXBIWXMAAAsTAAALEwEAmpwYAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAHkLSURBVHgB7P0NkFzlmR983wKMjQQ2a0A2Wst8OA/a5SMLBmIX1PtIBCep+AFlnZiyxW6CSWHI4n0eHBCblF1loa0yFRtBYGuNA0sZk8RoXXZlE4n1VhLYFVWBYsPnho93pSfmw3pXtgXY2EgC24De+R98D2daM+rumZ6Znpnfr6qrv865+3RrdPr0v6++7kV7RxQAAAAAANiPAwoAAAAAAHQhTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANCVMBkAAAAAgK6EyQAAAAAAdCVMBgAAAACgK2EyAAAAAABdCZMBAAAAAOhKmAwAAAAAQFfCZAAAAAAAuhImAwAAAADQlTAZAAAAAICuhMkAAAAAAHQlTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANCVMBkAAAAAgK6EyQAAAAAAdCVMBgAAAACgK2EyAAAAAABdCZMBAAAAAOhKmAwAAAAAQFfCZAAAAAAAuhImAwAAAADQlTAZAAAAAICuhMkAAAAAAHQlTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANCVMBkAAAAAgK6EyQAAAAAAdCVMBgAAAACgq4MKAAAMkZ+/9nr53s4fl1d+9lp5/Y03CgzCu5a8o/zqke8qBx90YAEAYHIW7R1RAABgCCRI3rr9eSEy0+LAAw4oK5YfJVAGAJgkbS4AABgaf/PCTwTJTJv8baXqHQCAyREmAwAwNH6y+9UC0yntUwAAmBxhMgAAsGCofAcAmDxhMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANDVQQUAAJgz9uzZXbY/83R54fmd5cWRUyxesqQsP+a4svy448vixUsKAABMB2EyAADMAVuffLxs+vbGsv3ZZ5pAeSKnnfmh8pF/uLqsOOmUAgAAg7Ro74gCAABD4LHv7iiMlQrk22++sWx96om+1ltx4snl4ss/W448amlhrFM/sKwAANA/YTIAAENDmDzWfVvuKd+847Z9KpGXH3N8WX7scaNB8fbnnikv7Nw5cv70mOXS/uITF11Szl55buEtwmQAgMnR5gIAAIZQguTbv3rTmNtWX7CmnDUSDE9UbZwq5s3f2ljuu/ee5vqe3bvL7TffVF4ZOf/IR1cXAACYCpXJAAAMDZXJb0p/5Ot+//Oj1xMef2bt58ryY4/vaf2Eytet/9zoBH1x9bovlhUn6qMcKpMBACbngAIAAAyNBMFfa1UkJ0j+wpdv6jlIrutcve7asvyY40ZvS4Xy/ibuAwCAboTJAAAwRO7fcs9oRXFC4bUjofDixUtKv5pq5qs/P7puQuq7/3RTAQCAyRImAwDAkEiP49rvOM6/YM2E/ZF7kXU/8alLRq/f/WebVCcDADBpwmQAABgSjz74wJiq5LNXnlumKmPUdhcJqx/9nw8UAACYDGEyAAAMiW1PPTF6+SMfXV0G5bQzPzx6eWvrMQAAoB/CZAAAGBLfe/bp0cv9TLjXzYqTTh69vPWpxwsAAEyGMBkAAIZEbXERy489rgzK8mPeCqZf0TMZAIBJEiYDAMCQaE+Ot3jxkjIoi5e8NVb6JgMAwGQIkwEAYEi0A+Q9A6wgbgfI7WAZAAD6IUwGAIAhcciSQ0cvb2/1T56q9lhHHLW0AADAZAiTAQBgSPzaia2J8p58ogzK9ueeGb3c7p8MAAD9ECYDAMCQOKEVJt9/7z1lUP77dzaNXl5x0skFAAAmQ5gMAABD4rQzPzzaN/mF53eW+7ZMPVDOGC+OjBVHHrW0nL3y3AIAAJMhTAYAgCGRyfE+8tHVo9e/+e9va0Llycq6m769cfT6WasEyQAATJ4wGQAAhkjC5DpJ3p7du8uG9Z+bVKCcda4bWbddldwOqgEAoF/CZAAAGCKpTv7dtZ8bvZ5QuN9AeeuTj48JkuMzI2PWFhoAADAZi/aOKAAAMAQe++6OwpsyAd/Xbr5pzG1nrTy3rL5gTVNlPJ6EyPff++flvo7J+y6+/Aq9kltO/cCyAgBA/4TJAAAMDWHyWI8++EC5fSRQ3rNn95jbEyYvP/a4svyY45v70g5j61OPj6lEjlQiJ0jOxH68RZgMADA5wmQAAIaGMHlfaW+x+Vsb96k27mbFiac0QfJEVcwLmTAZAGByhMkAAAwNYfLEtj71RLl/yz1NtXJnpXKVSuQVJ53STLS34sSTC+MTJgMATI4wGQCAoSFM7k2C5Ree/2F5cefOZsK+Q0ZO7z/m+HLE0qUm2euBMBkAYHIOKgAAwJySquMVReUxAAAz64ACAAAAAABdCJMBAAAAAOhKmAwAAAAAQFfCZAAAAAAAuhImAwAAAADQlTAZAAAAAICuhMkAAAAAAHR1UAEAAGbFnj27y/1b7ilbn3q8fO/ZZ8qLz+8sM2Hx4iVl+bHHlbNWnltWnHRKOfKopQUAALpZtHdEAQCAIfDYd3eUhSAh8t1/uqnc/Z1NzeXZllB59QVrFkyofOoHlhUAAPonTAYAYGgshDB565OPl69suHYoQuS2BMmfuOiSctqZHy7znTAZAGByhMkAAAyN+R4mb/rWxrLp2xvH3LbixFNGAtwPlVNHQtyZqgze/uwz5XvPPV3u/tPNZfvIedvqj69pqpTnM2EyAMDkCJMBABga8zlMTkuLP77jttHrCY4vvvyzI2HyyWU23XfvPU3I3e7XfPHlV5SzV55b5ithMgDA5BxQAACAafXCSFCbwLbK5Hdf+PJNsx4kR0Ljq9ddW5Yfc9zobd8cCb1fmKHJAAEAmDuEyQAAMM1uv/nG0R7JqUhOeLt48ZIyLOo2HfHLNht7du9uthkAANqEyQAAMI0y4d7Wp54Yvb52yILkavGSJU2gXGWbtz71eAEAgEqYDAAA0+juP9s8evmslefO2CR7k5FtywR8Vbs1BwAACJMBAGAaPfrgA6OXV1+wpgy708780Ojl7c89M9qeAwAAhMkAADBN0uKiyqR7w1yVXC0/9vhyxFHvaS6nd/KLO03EBwDAm4TJAAAwTV54/q0g9shfBrRzwa+dePLo5e89+3QBAIAQJgMAwDR5sRUmLz/muDJXHLLkrQkC288BAICFTZgMAACMsXjxkgIAAJ2EyQAAAAAAdCVMBgAAAACgK2EyAAAAAABdCZMBAAAAAOhKmAwAAAAAQFfCZAAAmCannfmhsnjxkuby8mOPK3PFipNObs6z7WetOrcAAEAs2juiAADAEHjsuzvKfLNnz+7mvIbKc8Vc3e5enPqBZQUAgP4dVAAAgGkzV8PY+RgiAwAwNdpcAAAAAADQlTAZAAAAAICuhMkAAAAAAHQlTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0dVABAGDOuvXWW5vTYYcdVo4++ujR82XLlo2e57YTTjihAAAATIUwGQBgDtu8eXNz/vLLLzen/amhcg2cc/n8888vAAAAvRAmAwDMYRdeeGFTmdwtSI4s8/DDD+9zu0AZZt/1119fvv/975errrqq+bIHAGAYCZMBAOawNWvWNKcExQmicr5t27Yx5zt27GjuG8+uXbsKMLs2btzYnCL/b2+55RaBMgAwlITJ80g+DL766qtloTvggAPKO9/5znLQQf68AVg40roipzj99NP3uT/tMNavX7/P7QIrmH3tXxbky5/84uDKK6/0qwEAYOhI2+aJH/zgB2XXKz8rP997YFnoDlr0RvnJT39ajjryyHLooYcWAFjo6iR946kBNMPhhZ07m/Mjly4tk7H1ycfLEUvfU448anLr9+uF53eWF3f+sKw46ZTC5F166aXNFzs33HDDaP/zfPmTXxTkPgCAYbFo74jCnPbSSy+VnT/+adnzhu8GqgMX7S3vPOj1svx972sqlQFgIaqB1JYtW0ZvW7ZsWVP5WKVieZiqkx/77o4yaLfffFO57957xty2/NjjyvJjji+rL1gzqeB1z57dzfnixUvKINz9nU1l07c2jo6bbTpr5bnN9vXjkk+sLitOPLlcve7aMmgJuhcfumTMc75u/efK1qeeKP/mD2/r+XUcb5yZduoHlpVhVMPjdluaVatWlXXr1vniBwAYCtLHeWDPnj3lZ3sFpm2v711Ufvb6m6+N6mQAFqLxQqm0v9iwYUM555xzRm9bKG0uFi9ZUj6z9vOj17c/+3T57yMB7ob1j5e1I8Frv4HyN79+W0lFxj+//IoyVQmRN317YxMcJ0B+ZffukYD28eb2bPdHPrq657EuHtmeI496T5kOt3/1xnLEyNjt53zx5Z9ttrWf12+8cXhT/j/mVwSZjK9+CZRzfZQBgGEhgZwHfv7zn4+Ep/4pO70xEii/9tprBQAWmocffnifIDmT9CWMat+2kIKpQxYvaSp26ykB7bov3dTcfvvNN467Tipoa9uJzttTQfzKnl2jl3tddzyPPvjAm1XIH3+zSjpV09m+hMv33fvn+yyfx5to/BW/fkqz/kTrjLetua3eXpcb7/lE53NO2J3H7PXxJhqn23oLSf5f5kufdnuL/Jrgsssua0JlAIDZpDIZAGAe2bhxY1PVWOWn8QmlEiZHe6KvtLxYyBKEfvJTl5Tr1n++qa5dceKboeh4LSfOHwl2zx4JfONf/9+XjI7x6IN/2YTAtR3FeOt+Zu3nRgLe4yfcjvySKtvSKYFyZ1XyeOO3K6tT9Zvn0W6P8cd33NasVyW4zvOubSZqlfWikVNtB5LxPnHRJeW0Mz+83+d8959uatb50h/e1tyXHsoJ59P6oqrL5r6JxtnfegtV/t/m/2/9/1wn5rvqqqtG/z8DAMw05awAAPNE/Xl8lbA41cjt4KndL1kP1tIErwlVtz/7THM97S8Svp616tzyB7dvLLd9c1M5dSRQ/ebIbTXATX/ghKw55fJH/q83A9/7ttzTrJsANOvmvoz/lQ3771989sq/24S9OSVUnch446eyesP6z024TpZ/7MEHmh7KeS6pxk5w3lmNnWUOGQm02+Pe/tWbxjzn5ccct89z7rR5JOh+ceQ5ZJk8XoL0tPDI65uAeqJx6nrZvrpeXo9s60KW/7t33nnnmF8R5P/4RBNqAgBMN5XJAABzXKqNb7jhhmYyveqEE05oQqfOVhbtNhdZhjISoh5a9ux+MzRNBXHnZHKpkE2wmUA0LTJyXwLoVPO2lzt7JIB+/8j6tc1Eljn1zA81lbsJiSfqK9xU3y4qTQ/nhL9ZPyF0qpLb6ySUTVVxrVbO+AmJs20JfTsntMtj5r4sk+1+8/kd1/QqTjV2e508709edMnouLVi+8VMljeyTvOcR8LmhMz764/8vZEwPuF7XSahcQLq+jgTjfPC8z9s1quvXdZb8ZVTZnWSvmGR/6cJj9uta3I9LS9MzAcAzDRhMgDAHJZwKT97b/dSPf/888uVV145bsjUDpOFUONLyJm+vamKTbXs9uee7nndhKGpbv7eSPCcdXutrG1aOoyc0ubh/i33NH2U7x8JodNqIu01EnZnvATWbQlmJ2oFke2IF3b+sDlVr/yy2riG43FER0B8xC8n8eu3d3G2L4F4tve0kSA9IXUvk/MlPJ7MegtFvhRKhfL69etNzAcAzCphMgDAHJVguHOivVxvT9zVqd3mQmVy+WVI+8NyxNI3g8t2797lxxw/WiWc/r7dJLz9ww3XlldGxkwYu7yp6H1P2VqeKL2qEwRGwtW010jQOpkJ6Wq19f3jTOJX23sMWqqm0y7j/i1/PvI6vtkmo7NH8yDXW0jy5U8m5ktVcm1zUSfmEygDADNFmAwAMAc9/PDDZe3atWMm1OtlYq728iqTS1MBHO3J92rP33ZlbILdbrLM+0cC5Isv/+xoALr1ycdHJ7UbT8LrLHPa3/nwPqHpaWd86M0WFrvfakeRCuMaNndTJ/X7zNWfm9FANpXUdbLChPJfue6LzWvZbTK9ya630Ew0MV+C5tNPP70AAEwnE/ABAMwx6Y2casQaDCdY6pxobyLtKuaFXsmYEDeVv+3exKkqXv7LHsFVJr4bzysd1cIvPP98U5HcDm4Tiu7P4l9OdLfpWxv3ue/Rhx5otqP2GU6IfH9HMJ3n8K9+95JxJ+6r1cd3/+mmfdb549aEgoOS0Dvjtlt7ZJvzenZ7rInW29+EhAtZ/q/n/3z9Qij7guwTNm7cWAAAppPKZACAOaT9E/dYtmxZXz9xbwfQC6kyOcHv126+afR6+iCnZ3B6/NaJ5+KEk05u2ixk2QSaqVJOqNspPYbv+/bGZrn0+E0rirNX/t1y959tasLfTDC3bSRI/usuPZNrz+OEyQlj3//LCehSMZ0gul2Vm+3MpHjr/9UV5SP/cHWzbalcPmvkOYzXX7iOnaA2oWx9PnWdfquVE07n+f3xHUuaqukVJ52yz+Ol1cdjI9vetK345WuQ5/HJiz6933EmWu/qdRcWxpcq5PRRbre6SbVy/o/vr9UNAMBUHHjNiDIPvfHGG+UXv/hFc3rttdeG6pRtO/DAA8ugvPTSS+XVvYMbb744aNHecughby/veMc7CgDMBwmK7rjjjtHr/QbJVX4W/6lPfaqccsopZdj84Mcvl0FLVezb3nbwmNuO/tX3lX9y4UVNeNn2/mOPb/onJ8x89KG/LIsWlaZtRYLeXxsJQd91+K80yyUATXj69P+7rbnt+P9jRXNbHue+LX9envirR0eun1z+0cfXNMejaWPRuQ1VwtX6mBkvIeqRS99T/umnLx9t+xB5nDPP/j/LD//mb5oezj/Y8Tfl7HPObZ5HldD8vSPPLc8vsl117Iz705+8VP7O2f+fMevk9TlyZJk897afjBxjdj7nPIeMc8Qvl29e24MPLief+sFmmYTqeb55/nn98hrleWQ7Rp/vyDgZO2F+qo+zrVkv4+R5tdfrtaVHv9777vnxRUq+EDrnnHOayfh27drV3JYWOKHlBQAwHRbtHVHmmZ/+9KdNwPqL1/eWvWVRGTYHLHqjHHzQQeW9731vOeigqReHP/vss+Wl1w8ujPWORa83HxQOP/zwAgBz3fr165v2FlWCovRInW/VxY99d0eB6XbqB5aV+STVyOmhXoPkOP/888u6desKAMAgzbswOd/I73zhhbLr9beVN4YwSK7ePhJ0Hnbwoqai6IADpta6Wpg8PmEyAPNB7YW6bdu20dvmc0gkTGYmzLcwucqvF9p9k0844YQxvZUBAKZq3k3Al6rkPW8cNNRBcvxs74Flzy/eGP05GgBAp/RB7QyS0wtVtSEwnquuumpMv+TsO37rt35rzMSbAABTMe/C5J///Ofltb1z42llO7O9AACdEv4kFOoMkk2sBexP534iPdLzpZRAGQAYhHkXJgMAzHU1SG6HP50VhwATyb4i+4xKoAwADIowGQBgiIwXJKetxZo1awpAr7LPuOaaa0avC5QBgEEQJgMADInOIDmTZmXyrEy4B9Cv8847b8wEfAJlAGCqhMkAAEMgvZEvvPDCfYLk008/vQBMVvYh4wXK7X7sAAC9EiYDAMyyhDoJd15++eXmeg2STzjhhAIwVdmXCJQBgEEQJgMAzKLOIHnZsmXlzjvvFCQDA5V9SvYtRx99dHM9+xyBMgDQL2EyAMAsGS9ITvVgDXsABin7lltvvVWgDABMmjAZAGAWpDfyVVddJUgGZtR4gfLatWtNygcA9ESYDAAwwxLaXHrppaPhjSAZmEmdgXLtoSxQBgC6OagA43rttdfKT3/607Jnz54yFxxwwAHl4IMPLocffng56CD/tQGGVWeQXCfbEyQDMyn7nOuvv3601U4NlO2PAIbTz197vTz/k93lJ7tfKQzGgSM5yiEHv628992HlYMPOrDQG4kTjCNBcg6o97y+qPx879wo4F9U9paDf/ZKefXVV8t73/tegTLAEBIkA8Mkk/JlHyRQBhhuCZL/944Xys9/8XphkF4vr/zsF2XXKz8rf+tXjxQo90ibCxjHCy+8UHaPBMmvvnFgeWPvojlxen3k9MrI9u7+xd7yox/9qAAwXBLUjBckJ8wBmC01UM4+KWqgXPu5AzD7fvCjlwXJ0yhh/fd2/rjQG2EydHjjjTfKK6++Wn72xtz8RioBeKqTARgeCWU6+5GuW7dOkAwMheyLMiFoJVAGGC4/2e0z/nR75WevFXojTIYOCZP3lkVlrtpb3nwOAAyPG264oWzbtm30eoLkVatWFYBhcd5555Vrrrlm9Hr2Wdl3ATD7XvcZf9p5jXsnTAYAmEa33npr2bx58+j1tLo4//zzC8CwSaCcfVSVfVcm6QMAqITJAADTJEFyTlVCmnZQAzBsOvdTGzduHLMfAwAWNmEyAMA0SEVfO4BZs2aNILkHBx7g8JTpZab27jp/QdH5CwsAYOFytA4AMGCdvUY7J7diYke9a0mB6fSuJYcUuktv99NPP330emfvdwBgYRImAwAM0Pe///0mOH755Zeb68uWLdNztA9HHb6kHPL2gwpMh0MPObi8992HFXqzYcOGcvTRRzeXs09bu3Zts48DABYuR+oAAAOSsCU/D69hS4LkW265ZTSMobu0uVjxvqXlRy/vKbte+VmBQTn0kLeXdx+2uNC7ww47rGlxceGFFzb7tx07djRflmW/lvsAgIVHmAwAMCAJXdpVe+2qPvqT0E/wB7Mv+7Dsyy677LLmem3jkzYYAMDCo80FAMAAJEjeuHHj6PVU76VXMsBcl97J7b7vmYyvvb8DABYOYTIAwBSlUi9hcrVmzZrmBDBfdO7X0gv+4YcfLgDAwiJMBgCYgjrhXpVq5PZ1gPmi8xcX69evNyEfACwweiYDAEzBNddcM2bCvVTrAcxX2cfViUYzIV/2gZmQD4C55YXnd5a7v7OpPPrgA6O3LV6ypCw/5viy+oI15cijlpZ+jDfeoE1l+xgcYTIAwCSltUX7Z96ZkMqEe8B8ln1cAuQ6IV/2gdkXJmAGYG5I8Hvd+s+VF0fO2158vpTtzz5Ttj31eFm77tqeA9s63mlnfrhcPbLedHll9+7y6EMPlA0jj9XP9jFYwmQAgEmoAUqVICWTVAHMd9nXZZ9X94E5z232gQBzw+ZvbdwnSG5LOHz7zTf2HAxnvLNXnVtWf3ya5ww5qpTlxx5Xyt7S1/YxWHomAwD0KT/vTmVeVYMVgIWi8wu09E9++eWXCwDDr5dWFNufe6b0KuN95KOry0zJY/WzfQyWMBkAoE+pwmv3SW4HywBz2ebNm3sOhbPvO+yww5rL6Z+cQBmA4bdnz+7uy+zuvkx7vMWLl5SZkt7J/WwfgyVMBgDoQ4KWnKpPf/rTQ9knuYbdAL3asmVL+aM/+qOya9eunpbPvm/Dhg1j1t+4cWMBYOHpJaBmfhAmAwD0KAHtDTfcMHp9zZo15fzzzy/DJNuYibHyE/QzzjijXH/99X56DnS1bdu2Zn+RcLifL8jS6iL7wqr9yw0AhtOKE0/uYZlTSq8y3tYnHy8z5b4t9/S1fQyWMBkAoEf5SXcNZtPeYpB9kjPuIAKYBDmrVq1qqqevuuqqZtz8DF2gDEwk+57sL3I64YQTSr+yL6wBdPY1Wv8ADLdPXnRJ12UuvvyK0quMd/tXb2pC3umW0Pqbd9zW1/YxWIv2jijzyLPPPlteev3gMhccvOiNsvSd7yhHHnlkmYq59Jxn0jsWvV7e++7DyuGHH97Xeq+99lr53v/vb8pPX39bmasOP/Dn5dhjjy0ADE5+up2qvSph7aDaWyTIqcH0okWLmtYZk614TjXyQw89NDpJYE65fu+99475OTpA1P1PTnW/k0A4lcrpg9zrvujhhx9ufhVRJZhuVywDMHmPfXdHGbTtzz5dvnbzH5Ttzz095vZU/K6+YE1P1cvjjffi8z8shyyZnv7Jr+zeXZYfe/yktq8Xp35gWaG7gwoAAPuVsOXOO+8cvd6uwhuEVBNfeOGFTfCSnqMJrWsw0+/jZPlsb8Y877zzmuupolaZDHTKfiGhbwLj/KIh+518UZYvn3JfqpR7DZNru4vaM7n+SmIYe8oDUJpQdt2Xb2x6HdfJ7DKx3WQn0htvvEGbyvYxOMJkAIAu2j1AB93eIuMmtKkVfAlfcspjJky+5ZZb+gpjEgxlvUMPPbSsW7euuS3hToJlgLb169c3bXBSgZwvtLJ/y74iwXAm4mv/GqMX2TfmC7G6X8svI7IPA2B4JZwdZEA76PEYPnomAwDsR6r0cqoGEYzk5+Pt8RO+tB8jEsr0GiTXSfdyniA6bTIS5CQcOuecc/qqLgQWhuwjtm7d2uxjsn/Iry+yz1mxYkXzZVa/X2RFgul2v+RUOmf/BgDMHyqTAQAmkLAloUpV20ZMRW1fkT7GUVtm5HE6W1v0+lgZq/Y2bfc+TbicCuUEPABt2S90folVJ+JLRXJ7/5N9U6qWe9kndba7uOGGG5rb7IcAYH5QmQwAMIGEIe32Fu3JpSYrocpf/MVfNOOuXr26Oa9VgQlb8hidAc/+JIR+5JFHmvVzStVzHTfBjwAH6EV7Ir78miEBckLl/Loh5/ULsF5kjLrvyRddNVgGAOY+YTIAwDjqJHZVWkcMaiKphCwZK2MmdMnj5LZUBOan5XfddVfPYyV4rsFNAqANGzY0Y6fFRQIgE+8B3dSJ+PKlWb6QSoBc90v1i6p+WuVkvSuvvHL0ervvPAAwtwmTAQDG0Z54arp6DmfMGrKkmjiVgAmC++3L3BkYp29ygpyMu3bt2gKwP+lrXHu5J1DetGlTsx/KF1WpLO7n1xJV9m/5JUbV7qUMAMxd8y5MPvjgg8tBi94oc0G2M9vLcDnooIPKQQcsKgcs2lvmooNGttvfFcDU1EnxqnawPFV1srxUDudxEh6vW7euqVJO2LK/0CahcZZpV/hlnATG7dtSRZhAKOMmCALYnwS/ab+TADn9jlNZnP1N9n3ZX7UnDe1Hqp2rfFmWEwAwt827MPmd73xnWXzA60MfBCZIXvy2A8rixYsLw+fwww8vSxa9NucC5Wzv4gNea/4fADB5g550r632JE0onImp2n2T87ipKp5IAp6ExHXZSPCTbUyonNAnP0+vFYEJgLI8QDft/uoJfbNPiQTMua9ej17D5fyqI/uoav369QUAmNsOKvNMZix/7bWREPCll8ov9h5QhjEKzFa9420HlqVLlzZVsAyfhLFvvPFGWfTST8rrexcN5d9Rp0Ujp4MPLOVXDv+V5v8BAJOTyuBBT7pXJaBZsWJFE/Tmcs5zSvhbQ+ZuskwC52xXeivXn6InsEnAU/sxZ/wE1u1gHFhYsh/I5Hdbt24d7auefcf+viBLNfK9997b7D9qm4qcZ93Ifqb9y4pusn/KviqVznUyvnbADADMLfMyyUxVacK0n//8500gOFU5mMrPviLVPitXrixTkQD5He94R2G4DfrvqHrkkUeaPnRxxhlnNNVkg5JK9wMO0AodYLI6J91L4DHIquS2hCsZPyFNQpl2VeD+pP1GKprTE3nXrl1NUJMwuo4VCXvyPBIGTdf2A8Mt+4Aa3KbdRMLc7D8SJn/jG9+YcJ+T0LhO6tmW61k3k3ymlU6v+5asl22o+9ac5/i3130eANPjhed3lru/s6k8+uADo7ctXrKkLD/m+LL6gjXlyKOWln6MN96gTWX7GJx5Wxbb9L0dUNXv3/zN34z2TczBlarPhWOQf0fVT37yk9G/p/wtffKTnywADIfOquRBV88l7K1tJ/JekJ7GkZCn18dKEJNq5BrkpEq5Vv4lYM71HK/0O4kfMH9kX5Z9Qjv0zXn2DalWrl9mjadzstHsn7LfyTo1ZK49lXsNhPNYdf+a9RJy9/JLDACmR4Lf69Z/rrw4ct724vOlbH/2mbLtqcfL2nXX9hzY1vFOO/PD5eqR9abLK7t3l0cfeqBsGHmsfraPwdJjAQCgvFmV3J78LhPiDUrCk1QLJ0xuV/MloMl9++uTPN52tr/YznipOkw4rdIPiOxb0q5ivOrhVAX3OhFebZeT9jz5ZV32MRk359lf5nJtf7E/WT7hce2ZXCum7bMAZsfmb23cJ0huSzh8+8039hwMZ7yzV51bVn98mtsYHVXK8mOPK+lF2s/2MVh+Dw8AUN4MX9pVyZ3VeZOVMdNfNIFMJturgXUqBmtlX8LgXiUI6uyDnHGyzbXNBbBwZX+QFjgThby5r5cQN4Fvwt/su9LaolYjZ/3sgxIO9xIkV3Vi0LqNGR+A2dFLK4rtzz1TepXxPvLR1WWm5LH62T4GS5gMACx401mVnHHrZFVpa1F7GUdC5H5/6p3lU+WcYDphTE4Zc1DhNzC31aA4ge148iuG7JO6yS8m0lu5BsCpUs56+WVE9kP1y7d+tPd32XdNtI0ATK89e3Z3X2Z392Xa4y1evKTMlPRO7mf7GCxtLmCGtQ+8a99MAGZXu9J3kFXJkWCntrGok+3l8TKRVbv38f4kxKnBSyqTs17GSagcE/2cHViY0kIiE3XWvuxttcq4m/Y+JfuYTEqeL8La1cX9yro5ZZ+mdzLA/DLTgTKzR2UyALCgTVdVcoKSBMYJYVKB3P4yMeFJr0Fy1stPzRMi156juS2Bd8bNqZ+fmgPzXw1o8wuGtWvXjmkp0W+f4lQyJ/xtVylH2l1MJlBWnQww+1aceHIPy5xSepXxtj75eJkp9225p6/tY7CEyTDD2mGCSUcAZl87SB5kVXICmAQvDz30UFOZnHHbFdC9VhKnt3IC7oyRQCcVh/mZeYJqgImkKjlVyCtXrmz2RQltsw9Kq4raeqcX9Xi1nmfflv1ZQuGE1Z093Lup1cmhdzLA7PjkRZd0Xebiy68ovcp4t3/1pibknW4Jrb95x219bR+Dpc0FzLB29YWfJAPMrumqSs64aUFRg+mELrlcJ+HrtSo56mRZGTMhTiqdc33Hjh3Nbd5LgInkVws51V9K1Ak/c559SfYv+YJqfxL81j7vtU/7okWLmkrlyLjtgLgX2SfWL8QyXrZBkQXAzFl+7PFl3ZduLF+7+Q/K9ueeHnNfKn5XX7CmHHnU0r7Gu/oLX2zGS9B7yJLpaXfxyu7dzWN95urP97V9DJYwuQftAxs/w2Kq2n9DDpoBZleqhusvRgZZlZwK4gS/keAmEvrWXsf7kyq/djCT9hZpbZFAKIFLxsk2J8wRJAO9SGCbfUgNhSNVy6lQ7hYmR22lk19KZB+V/VvOs3/L+vWXGL3KsnVfpncywOxoAuUv39j0Oq6T2WViu8n2PR5vvEGbyvYxOMLkHuSnpJUwmamqkyWFCfgAZlf759mDqkqOBMAJSzJ+ZyVyt8A6y9VJrmrFX8ZL2JJjklQSJtCpITVAN/mCq3OfUX/10K/soxIgZ/+UcR955JHywQ9+sPQrQXaqo0OYDDB7Es4OMqAd9HgMHz2Te9A+yMpB17A54IADyqLCXKHNBcBwSIXwdFQl50vDBCwJftOzNCF1QpKEJu2++ROpvZVz/JGwJtuZ9RMgr1ixonkfyVjpoQzQi1QW119LRPYjCZf73Y9kX5n9W+RLr0zul+uT2R/lS7L6OSvb094+AGB4qUzuQbt6NP0Jh83BBx9cDnzttfLaXt8NtB18wBvlHe94Rxkm7ark+nNBAGZHe9KnXn7m3Yu0o9i6dWsTkCRgSQic0xlnnNHXJFUZI0FPKveyfsZKsDyo7QQWlrovyWeZ7J/uuuuuJszttxo4AXLWyRdbdYK/ekybfWoKJXoNlrMd2afVfWPW9yUZAAw/6WMPhr3NxeGHH14WH/B6OWDR3kIpi0Zeh0MWvV7ecfDbhi5Mbn8ZocUFwOxJONv+gm8QAUYqiPfu3dtUENfeyKm0qwFLgpdefpGSY40bbrih2b4sn3HynpEwuR2AA/Qq+5Lajz37k4TACYVzW34NkbC5Wz/3Ok72cQmiI+PVX1HkS7CE1P18cdaeeC/75Vr1DAAML5XJPWiHfnWSiGGaOC2B6XuXHlV+9KMflVd/8XpZ6A5MmPyOt5elS4dvZs/2AbLKZIDZk8CjSigyiLZD6RuaCuQqY6YdRa8TXFU5xkjwnJ+P117LCX0S+PQS9gCMJ/uS2s4nn2myb0oYnGrj7Hdyns893SbSy7L5Ai5j1BYX+XyUQDmPcc4554wJibuNVXvCR/Zx/UzkBwDMPGFyj+psw5HzYQqTY/Hixc3p1VdfLbMhH8rrB9wcpNZqhdmQth/pIz2MtLkAmH15H2+HsglUBiH79YceemjMe2C/IXWCmQQpOSVATjjzjW98oznuqKEywFTViuT2PiWtdeovIRIsZ1+2v17yWTYhcsLknPJLjITI2Rf283kp69Yw+d577x26wh0AYCxtLnqUvmBVOxAcNqlSno1TqqKffPLJ5pTLs7UdOQ1rkBztymRVFwCzI4FvlX3xoL7cS/CS/Xx7or0ENitXruxp/doXefXq1U2QkzZbGTMVygCDlP1gZ3uftGPL55wEzLmv26Sk7cA3+9J6nFsD5V7VL9AiQbJ2PgAw3ITJPWoHf+kHBv3qbHGh4gJgdrT7eU72lzQJixP85ufcNfzdtWtXM3bOU+2c+2pA3IscaySEyTYlyEmFXir/MoZwBRikVB+354JJiJz9V45Pc95La57sq2pf+Oyn6niT6UHfXsf+DgCGmzYXPWp/u25iCCaj3Z9TiwuA2ZH38Fo1nDClW+XdeLJ+7WGcSaxq24wEyOl1XE+TkZ+Z10rkOkZCHl9AAoOUL8Cyv6mhcq1I7re/e4LnnDL5aMacrATTGSfbklNt+QMADB9hco/a4V8OtvTyol/tn1VPJrwAYOraX+x98IMfLJORwKM9qV6OBxLKZN+eIDitsfrplZwg+o/+6I+a7ckYCZHrRFa57gtIYNCyf8kEn9nXJFDetGlTsy9LiJtq48h+rtu+LPdP9suzts6J+HIuTAaYXi88v7Pc/Z1N5dEHHxi9bfGSJWX5MceX1ResKUcetbT0Y7zxBm0q28fgCJN7lAOcdi+w/PR0NieZY27prIRzcAww8zon3uu1/URb1k9f0fHCk4S+OTa48847m3C5VwmME0DnvSJBdb60zlg33HBDc7swGZgO+XyTauK6j0m/94S4df+VSuXcNlP7oPZEfLVthuIdgOmR4Pe69Z8rL46ct734fCnbn32mbHvq8bJ23bU9B7Z1vNPO/HC5emS96fLK7t3l0YceKBtGHquf7WOwhMl9aIfJ+cZemEyvBlEJB8DUtH8hknCkn+rhKoF0DaXH+5VJxq1Vfb1IWJLl80VjKp1zqj85z3uHIAWYTjUozj4t+7a6D0qQnPsSJqeCeSbUifhqkJx9YD9tNwDo3eZvbdwnSG5LOHz7zTf2HAxnvLNXnVtWf3ya99tHlbL82ONK2Vv62j4GywR8fWiHx+1JJmB/Oivh8pNBAGZe+4u9yQYUCVjSJzkVxKlsrr86qXJ80GsAnBA57wlZJ9WBmcgv49VfQ6X6eTKBN0C/sn/M/i2nHLfWXu0z/XmnPRFfP1/MAdCfXlpRbH/umdKrjPeRj64uMyWP1c/2MVjC5D602xPkwMpMw/QigUOVvx8/VwaYeQlp2xPonnHGGWWysh9P2JJ9eqqT636+PkavXxqm4i/rJkjOeAm4E1D7shqYaTU4rq0v6v6pHe7OhM7inc4v7AAYjD17dndfZnf3ZdrjLV68pMyU9E7uZ/sYLGFyn9oHVAmTfeBjfzqrkrVGAZgd7Qq3vJcPouK3VvDllKrihC+9jp2+ywlt2ssmTE5A3a6gBpgJ2f+sX7++qUiuxQ/ZP2U/N5PqLzOq9nE0AMOtl4Ca+UGY3KeEgfXnq6qT6aZdlZzK9vH6awIw/TIpXrVy5coyKAlbalVxAuJeg5ccS+zatWufqju/XgFmQwLcK6+8cvSXEpmEr5+JRAep3YZImAwwPVaceHIPy5xSepXxtj75eJkp9225p6/tY7CEyX3Kh7/2AY7qZCZSq9WqmZq8BICxUmlXQ9u8j0/HF3s5NmgH1t3U44m0tajvFdnOjDHTPysHiOx7sg+a7eKHBNu1eKezRREAg/HJiy7puszFl19RepXxbv/qTU3IO90SWn/zjtv62j4G66BC3/Lhr854nCB57dq1gkLGyN/GDTfcMHo9fzMmUQKYHe0v9gZZlTwZeX946KGHmp7N9b0hX0zn5+WpSk5VoPcLYCFLkJxgu+67Eya3W18AMHXLjz2+rPvSjeVrN/9B2f7c02PuS8Xv6gvWlCOPWtrXeFd/4YvNeAl6D1kyPf2TX9m9u3msz1z9+b62j8ESJk9CnZgi1USRA5y0M5jpnmIMpwQF+VuoFetpb+FvA2D2dPZLni31/WHFihXNcUOq/3JdJTLAWGktWMPknDuWBhi8JlD+8o1Nr+M6mV0mtpvsRHrjjTdoU9k+BkeYPEn5djwVRbVncu2N60BnYatBQfvn1Klarz/VA2Bmdba4mM3gNj8fTw/SbEO2KccOmbgvfUoFygBvyS81ss9OcUZtdaE6GWB6JJwdZEA76PEYPnomT0E+ELYPavKhsD3hGgtLDnLbQXKsW7fOz5UBZtGwtLjIe0Qm3KvHDXlvyHvEpz/96XL99dfrCQrQ0vnln30kAAwPYfIUbdiwYczM6wmT0wKjc3Z25q9UTOTfPW1P2tVvCQlUmgHMrnYAMdP75HaQXXsld37pnFYXqVhWcQcwVlpdVO39KQAwu4TJU1TbGLQD5bvuumvM7OzMXwkpLrzwwjHhQHok529itmfCBljoEuCmzUXMRouLTMRa++cnFMl7RaqT09qiblfdNgDGqq0uIvtzxToAMByEyQOQg5xUFaWHcrVjx45mZvZ8YEyo7OBn/kgwkF7ZCYvb1ciRyrLOLxcAmB3tifdmo/I37wXtyuh2a4u0ysovmWrYDMBY+YzVPqZu79MBgNkjTB6gfDDs7JFbQ+UEj2vXrm2C5XY1EnND/s0SICc8Puecc5r+lu0QOQe7+fdPkKxHMgzGY489Vj72sY81/+fuuOOOAv1qBw+z0S85ldDj9fnMMUGqlPPeoSoZYGLtVhfCZJg5jsOB/TmoMFD5gFg/JHZWJOcAqB4E1W/ac0pbhASQ9UNlTgLJmZOqsPzsOOcJ/+us0bmcEDmXJ6ocy79VKtJzEgjA4Dz77LPNwetLL73UXM++M7flCzvoRfbd7SD3jDPOKDMh7xf1/SABdr6EzJeNndsWnbfDIORvMC3XDj30UC23mPPypVwKcyL79PY+FpgejsOBboTJ0+TSSy9tTgmUU9HaWY2cA6EcEPUzM3G/AXNC6sgBVz5Q5HoNsTt/NjasatCbSYtyXsPedvjbj7reVOXn0jm4TbWEA1oYvBy01gPYKi0BwoEsvcj7RpV99kx9SZv3/AR5V155ZfM+kfedBMr1/SpBcrYl9wmTmQ7p1V3n7cjxZ45HHaswV+VvN/vw+pnp3nvvHVOtDAye43CgG2HyNKuVynUW9xwA1W/V+9Vv3+Vuy7ero/Ohdhhmks/rkjevfPjJ6zQsLUHywT+vTw2RfSiD6XXssceOe7sDWXr1yCOPjF6eyYn3EtzlPSPtkPJ+VoPs/O3mi13vH0yn+su4Kl9u5NhTGy7msnaYnM9TwmSYXo7DgW4W7R1RmHHtNgo5r60U2pVLs6FW2/b7wTsfXnKKWpXdj3zgzgeefiq1B6VWbke75UhtQdKeSRqYOZ/61Kcm7NGWg1kHsuxPfp5Zv7jNJLkz/WucvI/XYK9OzgrTKX9zE7W1yHFMquW1vWAuyuelCy+8sLmcv+W/+Iu/KMD0chw+fB77bn+/yu7FC8/vLHd/Z1N59MEHRm9bvGRJWX7M8WX1BWvKkUctLf0Yb7xBm8r29eLUDywrdCdMnkP6DZhrMJ0P0+0+wDnPQdn+qqMToma2+V4/dEwmTM7jJ0DOaX/bknB3xYoVzXm2q1Z2tU+9UhUGc4sDWSYjX0ymtUTkfWPTpk1ltmRb8reaQDnbpDqU6ZBjuxx7dTtWzBwP2qswF7W/IMyXc8Pwi0qY7xyHD5dBh8kJfq9b/7ny4sj5eBLUrl13bc+BbR3vtDM/XD7y0dVluryye3d59KEHyv1b7ulr+3olTO6NNhdzSL8fQLst3+7b3NlSIqFzJrv4oz/6o75C5V7VD9fjfeip7SRyUhUMC9vXv/715ny8A1k/tWMi7V+5fPCDHyyzKe9lqU7OF64J8RKCeF9j0NJWpR5T5e+rhm7111f1Pm0vmKvyy8n8/Ub28cJkmH6Ow+e3zd/aOGGQHAmHb7/5xnL1SGDb63hnrzq3rP74mjKtjipl+bHHlbK39LV9DNYBhQUrHzDqBED5CXA+7NZej1UNlVNNNajWG/nA0zletiWPnZ+t5QNOLucg0QduIAeyF1100bj35UC2zvIOVTtMnsl+yfuT97W813pfY9DyRUXahVXtX4clVM7fXbsoIMd2OQ5r91aGYdduVTQbbfFgoXIcPn/10opi+3PPlF5lvOmsSO6Ux+pn+xgsYTKjEiLnA0g+XOQbxnaoXH8yPJUJ8RIep99ZrSqIGiLnJ8hmGwcm4kCWXtVf3VSq15jPclxW24xFjqXSyqJ9DLdr167muK4dMtdigfa6MMzaXwxOdjJzYHIch89Pe/bs7r7M7u7LtMdbvHhJmSnpndzP9jFYwmTGlQqWWqlc5YNHwuDJVLLUXn7tMDoHhUJkoFcOZOlFZ5Ds/YX5KsdW7f7Hqdysx22Za6LaunVrc14LBtpBc8LkHNvN1sTP0Kvsy9tfDqpOhpnlOJxe9BJQMz8Ik9mv+rPc9gePvFH0EyinciAfdtofVDLuhg0bfMgH+uJAlm7SD7Zq/ywa5pv23BOZaDJtxKr2cVv7+Cu3J0But73IF/11wkoYZsJkmF2Ow+eXFSee3MMyp5ReZbytTz5eZsp9W+7pa/sYLGEyXeXDeD54tD+Y3HDDDT23vMi67WU7f2oJ0A8HsuxPrcKMYemXDIOWY6t2mHbllVeOOU5LuFy1/09Elhuv7YW2AQy7dpjc7hMOzBzH4fPHJy+6pOsyF19+RelVxrv9qzc1Ie90S2j9zTtu62v7GKyDCvSgVrLkg0cqXPKBY+3ateUb3/jGfquLU8Hc7pGc9dvVMACTYXZpxpP3p/aXl/olMx8lRO7sk9z5xUk7WJ7oy/+6Xn6BlnO/FmPYpcAlf6f5HFI/j/i7hZnnOHx+WH7s8WXdl24sX7v5D8r2554ec18qfldfsKYcedTSvsa7+gtfbMZL0HvIkunpn/zK7t3NY33m6s/3tX0MljCZntVAOb31cvBWJ29Ju4rx5CCv/WEnE8KoSAYGxYEsnR566KHRy4Jk5qt2O4v8nY93bNVu8bK/fshZru4vYdglOM7fbK3KT1uj8847rwAzz3H4/NAEyl++sel1XCezy8R2k51Ib7zxBm0q28fgCJPpSwLlhMe1t15+YjZRz7IEye1efoJkYNAcyNL2yCOPjF7W4oL5Kn/bqTbOsdVEQXDuqxWc9aSCk/kgf//1s0e+QNxfmJyil7R5ydwtvmCEwXMcPn8knB1kQDvo8Rg+eibTtxyMpcq4yoFaZ5+9hMjtSfo+/elP+xADTAu926jalckm32O+ypfzt9xyS9NqrN3OolMvrS5grmnv2/c3CV/a7OWzSP722y33gMFyHA4Lk8pkJiUfZO66667Rdhe53NZub5HqGH2SmW7HHXdcgfGojFgY8iVm/TVMvrxUhcZ81svf94oVK0ZD5Jz7P8F8kL/jbn2TO1vt+XJx+jkOZyKOw2F+EiYzKTloS3VyPVDrrExuV4elKhmm27PPPltgIjmQXblypdYH81h+ylwJDmDs/4N88Q/zRbe+yfl8Uj+baLU3MxyHsz+Ow2H+0eaCSUuYPF7rivRRbvdKVpUMDIN84GT+av/7qsCEsW0u9tcOAIZJ/lbXrl07prK4U3sf3/4iMdLaot1qLy1hgNnnOBzmF2Eyk1arkzu1+/LlG0iAYfAbv/EbhfmrHSgIk2FsZXL9kh+G3fXXX98UpiRMXr169bh/u+19fJatOttb5HPK/vqKAzPHcTjML8JkpmS8MLlNVTIwDK644orym7/5m4X5KT9nbn+RKUyGN38dVn9BVvvLwrBLr+8q7VnyWSIBc7ulXucXJfW+BMntX0dqbwHDwXE4zD96JjMldZKj8X4+mYM4fSuZKc8880xh4UovtjvuuGPc+zLD9I033liYv9rvQYJkeEu7t2yq91VpMuwySVf+btt9jzdu3Nj8RD7zsCRc7vz8kfPc1m5vkWXHa8fH9HAcvrA5DoeFR5jMlKWR/nhhsiCZmXTssccWFqb169fv9wD261//emF+a78Hee+Bt7TD5Jyb/Ii5IL98zN9qAuUaEKdKOe/3+Tu+7LLL9vnbbre7yIR8fh05sxyHL1yOw2FhEiYzZemLnJ+fdVIdBky3HMCmGmI8DmAXDi0uYHz6JjNXpYo+Vcof/OAHx7SvuOuuu8ojjzwyZl+f22oVc34ZmbAZmH6Ow+e+F57fWe7+zqby6IMPjN62eMmSsvyY48vqC9aUI49aWvox3niDNpXtY3CEyUxZDtpywNf5IUV1GDCdHMBStSuT2/02YaFrB27tSSphrkiF8RlnnLFPlXJOVbufctpbaOcC089x+NyX4Pe69Z8rL46ct734fCnbn32mbHvq8bJ23bU9B7Z1vNPO/HC5emS96fLK7t3l0YceKBtGHquf7WOwhMkMRA7y2n3KQnUYMF0cwFK1g+T65SbwpjoJX52AL+f6yDLX1CrltK/Ie/9EVfb57KG9BUw/x+Hzw+ZvbdwnSG5LOHz7zTf2HAxnvLNXnVtWf3xNmVZHlbL82ONK2Vv62j4G64ACA9D54V1VMjBdHMDSlnCsvgd574F9tf9fjDfHBcwVCYtTvHLppZeOe/9ExwbA4DgOnz96aUWx/bneJ9fMeB/56OoyU/JY/WwfgyVMZlqoegGmgwNYOmWSpoQLOV111VUFGEuYzHyTMDn7/HYxS27zyxSYXo7D55c9e3Z3X2Z392Xa4y1evKTMlPRO7mf7GCxtLpgWDuaAQXMAy/5434HxtcPk9mSVMJdln1+/SAztLWB6OQ6nFzMdKDN7VCYzLdKjD2BQHMACTE57DgthMvNNQmRBMkwvx+Hz04oTT+5hmVNKrzLe1icfLzPlvi339LV9DJYwmWmhzQUwKA5gASavTsIXdSI+AOiF4/D565MXXdJ1mYsvv6L0KuPd/tWbmpB3uiW0/uYdt/W1fQyWNhdMC2EyMAgOYAGmLtXJW7ZsaS6nb/J5551XAGB/HIfPb8uPPb6s+9KN5Ws3/0HZ/tzTY+5Lxe/qC9aUI49a2td4V3/hi814CXoPWTI97S5e2b27eazPXP35vraPwRImMy127NhRAKbCASzAYLTD5IceekiYDMB+OQ5fGJpA+cs3Nr2O62R2mdhusn2Pxxtv0KayfQyOMJlpIUwGpuLZZ591AAswIO1J+FKZDAATcRy+8CScHWRAO+jxGD56JgMwdGoFXScHsAD9S5hcW5ClZ3J6JwPAeByHA90Ik5kWJncBpuLYY4/d5zYHsACTkyBZdTIAvXAcDnSjzcUc8tprr5Vdu3Y158PmpJNOKr/7u787ev3QQw8tL7zwQplL3vGOdzTbDcy+VatWlXXr1pU77rijvPTSS80B7I033lgAmJyEyTVEznn2swDQyXE40M2ivSMKQ+/nP/95+f4PflBeeX1ReWPvojKM/uRP/qQ8//zzzeUlixeX3/rt3y5zRV7Rtx/wWnnnSJh85JFHFgCA+SQB8mWXXdZcPvroo8vmzZsLADA3PPZd81LNhFM/sKzQncrkOeCNN94oO3fuLD/9xYHl9TKcQXI89uTW8td//f9tLieQveC3/lmZS372+sHljZf3lIMP/ml55zvfWQAA5ovaNzn9kmvf5NpHGQAAeqVn8hzw6quvlld+8cZQB8lx1FFzv6L31b0Hlj179hQAgPmks2/yvffeWwAAoF/C5DkglclzoRfJ4sWLyzHHvL/82q/9evn1X//1Mhe9sbcMZU9qAICpOv3000cvP/TQQwUAAPqlzQUD89tzqEcyAMBC0w6TVSYDADAZwmQAAFgAEibXvsk5bdu2bUzrCwBg4Xjh+Z3l7u9sKo8++MDobYuXLCnLjzm+rL5gTTnyqKWlH+ONN2hT2T4GR5gMAAALxKpVq8rmzZubyw8//LAwGQAWoAS/163/XHlx5LztxedL2f7sM2XbU4+Xteuu7TmwreOdduaHy9Uj602XV3bvLo8+9EDZMPJY/WwfgyVMBgCABeKDH/zgaJi8ZcuWsmbNmgIALCybv7VxnyC5LeHw7Tff2HMwnPHOXnVuWf3xaT6uOKqU5cceVzKxWD/bx2CZgA8AABaIVCZXqUxOuwsAYGHppRXF9ueeKb3KeB/56OoyU/JY/WwfgyVMBgCABSI9k03EBwAL2549u7svs7v7Mu3xFi9eUmZKeif3s30MljAZAAAWkHZ1cm15AQAwFb0E1MwPwmQAAFhAVq5cOXp527ZtWl0AwAKz4sSTe1jmlNKrjLf1ycfLTLlvyz19bR+DJUwGAIAFZNmyZeXoo49uLidITqAMACwcn7zokq7LXHz5FaVXGe/2r97UhLzTLaH1N++4ra/tY7AW7R1RGGq7du0qO57/UdnzxkGF6XVA2Vve/fa95X3ve18BAJivbr311uYU5513XrnmmmsKADCcHvvujjJo2599unzt5j8o2597esztqfhdfcGanqqXxxvvxed/WA5ZMj39k1/ZvbssP/b4SW1fL079wLJCd8LkOUCYPHOEyQDAQvDwww+Xyy67rLmcSfk2bdrUnAMAw2c6wuQqvY7rZHaZ2G6qE+m1xxu0QWzf/giTeyOdBACABeb0008vJ5xwwmjP5ITL7Yn5AICFIeHsIAPaQY/H8NEzGQAAFqB2eLxx48YCAADdCJMBAGABSq/kKpXJqVAGAID9ESYDAMACtGzZsqbdRXXXXXcVAADYH2EyAAAsUO1WF1u2bCkAALA/wmQAAFig0urisMMOay6n1UVOAAAwEWEyAAAsUAmS260uVCcDALA/wmQAAFjA1qxZM3o5fZNNxAcAwESEyQAAsIClMvmEE05oLidINhEfAAATOagAALPq56+9Xr6388fllZ+9Vl5/440C0+ldS95RfvXId5WDDzqwQJWJ+LZt29ZcTquLdrUyM8f7AZV9NQDDSmUyAMyiBAdbtz9fdr3yc8EBM+Inu19t/ubytwdVwmMT8c0u7we02VcDMKyEyQAwi/7mhZ8IDZhx+ZtL9SNUCZLPO++80eu33nprYWZ5P6CTfTVQHXrIwYXp5TXunTAZAGZRKo9gNuRn9NCWVhdVKpNNxDezvB8wHvtqIH71iHcVptf7l/5KoTfCZACABUgFJJ0yEV9O1caNGwswu+yrgTjk7W8rK9531Mi5qc8GLRXJf2vZEXrU98FfIQAA0Lj00kvLZZdd1lxOmNzupQwAzJ43A+WlI18y7R056ac+CAcecODIaVGhP8JkAACgUauTa5uLBMoJmAGA4ZDw88ADxHnMHm0uAACAUe2J+BIm650MAEAlTAYAAEadf/755eijj24u1+pkAAAIYTIAADBGu7WF6mQAACphMgAAMIbqZAAAxiNMBgAA9qE6GQCATsJkAABgH6qTAQDoJEwGAADGpToZAIA2YTIAADCuVCeffvrpzeUEybfeemsBAGDhEiYDAAAT6qxO3rZtWwEAYGESJgMAABNKZXKtTo7rr7++AACwMAmTAQCA/brmmmtGLz/88MPNCQCAheegAgAwR+zZs7tsf+bp8sLzO8uLI6dYvGRJWX7McWX5cceXxYuXFGDwjj766LJmzZqmzUWsX7++fOMb3yiHHXZYAQBg4RAmAwBDb+uTj5dN395Ytj/7TBMoT+S0Mz9UPvIPV5cVJ51SgMFK7+S77rqrmYhvx44dTbDc7qcMAMD8J0wGAIZWKpBvv/nGsvWpJ3pa/tEH/7I5rTjx5HLx5Z8tRx61tACDkSrkhMe1Z3LC5PPPP7+pWgYAYGFYtHdEYajt2rWr7Hj+R2XPG7L/6XZA2Vve/fa95X3ve18BmAmPfXdHYXz3bbmnfPOO2/apRF5+zPFl+bHHjQbF2597prywc+fI+dNjlkv7i09cdEk5e+W5hfGd+oFlBfp12WWXjfZMzsR8t9xyS2HqvB8wEftqAIaJdBIAGDoJkm//6k1jblt9wZpy1kgwPFG1caqYN39rY7nv3nua63t27y6333xTeWXk/CMfXV2AwbjqqqvKhRde2FxOqJwK5fRTBgBg/jugAAAMkfRHbgfJCY/XfenGsvrja/bbtiL3XXz5FeXf/OFt5YjWcn98x21l61OPF2AwTjjhhDHh8a233tr0UQYAYP4TJgMAQyPVxV/rCJK/8OWbyvJjj+95jKxz9bpry/Jjjhu9LRXK+5u4D+hPeifXXskJktevX18AAJj/hMkAwNC4f8s95cWRQDkSCq8dCYUXL15S+pV1P3P150fXTUh9959uKsBgZDK+a665ZvT6li1bmhMAAPObMBkAGArpcVz7Hcf5F+y/rUU3WfcTn7pk9Prdf7ZJdTIMUCbfa7e7SHXy97///QIAwPwlTAYAhsKjDz4wpir57JXnlqnKGLXdRcLqR//nAwUYnM52F+1qZQAA5p+DCjBq0aKRb1gO8B0LwGzY9tQTo5c/8tHVZVBOO/PDZftzzzSXt448xtmrph5SA2+q7S4uu+yy5vrDDz9cNm7cOKZiGQAWivvuu69897vfHfe+s88+u3zgAx9o7s9y/+yf/bMyaLt27Sr/6T/9p3HvO/TQQ8s//sf/uMxH//7f//v93v/e9763/P2///dLNz/84Q/Lf/2v/7Xrv81f/dVfNcv2MuZ8JEyeAxYvXlzefuCPy6t795Y39i4qTJ+3L3q9HHzw4gLAzPves0+PXu5nwr1uVpx0cinffvPy1qceL8Bg1XYXCZHj1ltvLatWrRqtWAaAheL+++8v/+2//bfyt//2397nvlNPPbU5/9//+3+X//Af/kPfYfLatWubMPiss86acJndu3c3Y7/nPe9pTm0JVActwXVC1d/5nd8psynhbpVA/emnnx7zb5DXpRc/+MEPevq3yeP9r//1v4TJDK9Uyv7K4e8qr/9oJFB+48DyWhEoD9qisrccPHJa/LYDyrvf/e4CwMyrLS5i+bHHlUFZfsxbwfQreibDtEi7i1Qlb9u2rWl3cdVVV5VbbrmlqVyGtu0jXxwO4gvDQY0DMGgJca+//voyaAkwew0vE4bORNCZKuuEybOt/XqnsnjDhg3T8m/Am4TJc8Q73/nOJlTONyyvvfZaYfAOPfSw0dcZgJnXnhxv8eIlZVAWL3lrrD27hckwHRIa50PbhRde2ITJCZVToZxQGSL7+OvWf24kBH6mrDjx5HL1umvLZH1lwxfLow/+ZfPFY8YZ5HsGwGxIKJuK5hrMpvq4hsG5r7auyDIJlRMWd1Ye9ytjpro2OVMeqzN8TjVvgtm6TFp0pDK6Pu5Xv/rVZluy3HXXXVd+4zd+oxkjt2U7r7766jHjZfksk+eW8XI946WaO+tcfvnl5fjjjx99nrm9blvWm+rzjby+tZo62q9z53PPctmutAfJ6123bSJ1ndrmZKKx5wNh8hySP+CcAGA+ShhQA+WcDyocaAfI7WAZGKy0tUiFcq0EStuLtMBIywsWtnaQHFtbPfInY+uTb66f8TKuQBmYyxKarlu3rglM05ohQWeC1oSTH/vYx5ocqAap7ctT8S/+xb9oWjr8g3/wD5qQND2HE4T+u3/370aXyXwICXPrMgl40+s5y9TtqBlV+3LGHS9MzrpLRo7FE7LmuWWZBK9ZL8977969zXJ57gmx29sW9XEnK69rntPf+lt/a7QFRh4r29DZpuPLX/5ys4313yPrJezOv8dEY+cL9LQSqeskYM/5P/2n/7TMN8JkAGAoHLLk0NEwOT9fXnHiKWUQtrd6MR9x1NICTJ/0Tk67iy1btjTX169fX1asWKF/8gLWGSTH6gumNkFj1v/jO25rLguUgWFTg9K2hKAT9TpOiJrAsd2nNwFqTgkvE9TmvvTy7bXatd1DuKrVvdm2hMRpR1WD6QS3CZgTKKdaOPdn+XYFdG7P9qRSOduR67XNxWQnE0y1czt0zrZlG/7jf/yPYx432/Ynf/InUwpm06s6zynHJlX+XfK6dobJub29XELw/Hv8vb/398YNtHNfguS016jy75Q+1/VLgvnE7/kBgKHwayeePHq5Vp0Nwvbn3gow2v2TgemR6qoaHtf+yTln4ZkoSF798amFyR/56OoxgXQNlPfoiw8MgQSxNQyup9qmYjxnn332mDC2ThaXcSYrVcCd21DbL6TqN4/ZrnDO5QTECYojgWlC3rpMtqm2hpjKdnVKiN253Z3bVoP43DcVGbcdELdf586+z//kn/yTMdfz75Pl6uvTKSF4Z6Bew/uE2PONymQAYCicMBIm33fvPc3l+0fOp1q5Vv3372wavbzipJMLML3a/ZND/+SFabqC5NGxfjnOpm9tbM5VKAPDIgFiKmt7VXvtJixNi4hcn2ori7RkmKiCOcFpQtHxAu4EoFWqm1MNnDA06yyZgXZxNbBNBXCnqb4m7T7IeU65PtFz6uyPXNvOjhek1yB6ouOcQYbvw0KYDAAMhdPO/HD55h23NQHEC8/vLPdtuaecvercMhUZ48WRseLIo5aWs1dObTygNyeccELzoardP3nZsmVNGwzmv+kOkkfHFCgD80D686ZqOO0cEuamXcL/+B//Y0zLhEEbb8K9qC0cUmmbnr/ZpgS76TOcMHe8kHfQaguNibZtsm6++eYmsE8Fcdpa5HVOqJxWFJ0SAHc+XrdQOGOmbUenPM58I0wGAIZCJsfLT5c3ffvNUOCb//62suKkU5oQeDISSNex4qxVgmSYSQmOU5W8efPm5nqqkzMhX4Jm5q+ZCpJHxxYoA3NcJuBLS4l2uLt79/S17UngmWradhVyJNCuYWiqd3N/u5dwZyuIbtqBbNbtZf08/tNPP73PtmXdqVYm5zklpG4H1RNtU7ah3ee49qBOqN6pTj6Yf7Pp2O5hpGcyADA0EibXSfL2jByQbRgJBF74ZWVxP7JOwoR2VXLGBmbWlVdeORoep29yqn++//3vF+anmQ6SRx9jZHw9lIG5Ki0Vaj/jSHA5XguKBJZ1wrupSJiax8jEc+3HzCR3deLAhLrtx0pQmr7LndImIsvlVAPw9CaOr371q83tGTvzKfRSWZxexVmnc9vya6eMNxUJdds9j/PcxntOkSrm+txznirtBMWd7S+qTAyYf7N8MVBl7N/+7d8e8287XwiTAYChkerk3137udHrCYX7DZS3Pvn4mCA5PjMypgo1mHm1f3LOY8eOHXonz1OzFSSPPpZAGZijfu/3fq9pv/Cbv/mbTfiY4HW8FhQ1sMwyE00E14tacZyJ+NK2Io+bL3sTMtfHzXnC1zxWTllmvHYNCX8TKCeITgAbCY2zrQmmc3t9Pr30XM5jZls6ty2P3TnBXb/SRzo9qTNuntNll102bjuNyPbX555THj/V4xPJOB/72Mea55ptzin/VhO1vpjrFu0dUQCAWfHYd3cU9pUJ+L52801jbjtr5blNUDBR24uEyPff++ejk/hVF19+hV7JEzj1A8sKzISHH364+dBWpQWGUHmsufx+MNtBclvaG9WWF7H82OPmfMsL+2qYO1Kdm/YO+2ttMN4yuS39e6O2UhhvnCyXQDTh5kThbCppE+h2C287H3O85Wt7hzxetmWisbNcXaZz/Dp257rdtrPzsXvV7d+g3bKibld72Xo953mts42dgfBEj9FeZ3//RnOdMBkAZpEweWKPPvhAuX0kUO6sKkuYnHBg+THHN/elHcbWpx4fU4kcCQ4SJGdiP8YnoGAmZRK+OiFfXHrppc2JN83V94NhCpKr+RYo21cDMEwOvGZEAQBmxQ9+/HJhfEf/6vvKmWf/n+WVkbB4+3NvhRQJLn6w429GAuQnytP/79bmvlc6AucVJ55SPvu5a8oH/o8VhYm9992HFZgpp5xyStM3+Yknnmiup1r56KOPLitW+H8ac/H9YBiD5Mh7QFlUmveJ+OlLL5Un/uqR8ndG3lPe9raDy1xjXw3AMBEmA8AsEibvX3oop7J4xUmnJBdoqo9/8YtfjL/s4iXl5FNPL//005c3YcbiefqzskESUDDTzjrrrCZErpPwPfLII81tRxxxRFno5tr7wbAGydV8CpTtqwEYJtpcAMAs0uaifwkGXnj+h+XFnTubwPiQkdP7jzm+HLF0qUn2+uSn08yGVCdfeOGFo4HysmXLyi233NJUKS9kc+n9YNiD5Lb50PLCvhqAYSJMBoBZJExmNgkomC0JkhMoJ1iOBMrf+MY3ymGHLdwKzLn0fnDflnvK7V99a5LUYQ2Sqz++47Zy93c2jV7/5EWXlI98dHWZK+yrARgmBxQAAIAZlCrkVCPX8HjHjh3lsssuGw2XGW5HLl065nomTO2cLHVYZLsySWvbIdogAcCkCZMBAIAZd8IJJ5Srrrpq9Pq2bdvKDTfcUBh+6UecauQq7S7S9mLYAuWJ2nGcvfLcAgBMjgn4AGAWmYCP2WRSJ2ZbAuXIpHyRQDnVyZmUb6GZa+8Hwz7B3Vzq69yNfTUAw0SYDACzSJjMbBJQMAxOP/305rwGyk888cSY2xeKufh+MKyB8nwKksO+GoBhIkwGgFkkTGY2CSgYFgmOMylfKpOjBssLKVCeq+8HwxYoz7cgOeyrARgmwmQAmEXCZGaTgIJhsmrVqn0C5UzUt2LFirIQzOX3g2EJlOdjkBz21QAMExPwAQAAQ+HKK68c7aMc69evL5s3by4MvwS2szkp33wNkgFg2CzaO6IAALPise/uKAtVPvjfv+WesvWpx8v3Rj78v/j8zjITFi9eUpYfe1w5a+W5ZcVJp5Qjj1paFqpTP7CswLDJBHyXXXbZaIVy3HLLLfO+5cV8eT/Y9O2NZdO3No5ez/726nXXNvve6TLfg2T7agCGiTCZMV566aVy+OGHFwBmxkIMk/Oh/+4/3VTu/s6mGatY25+EygkdFmKoLKBgWKXdxaWXXtqcx2GHHdYEyu2q5flmPr0fzGSgvBAqku2rARgmwuRZkMD2Yx/72D63n3rqqeWKK64oxx57bOnHxRdf3PSYu+iii8pUfP3rX2/G+pM/+ZPym7/5m2Wm3XHHHeU//+f/3Dz+RM4555x9bpvs6wYwDBZamLz1ycfLVzZcOxQhcluC5E9cdEk57cwPl4VEQMEwW2iB8nx7P5iJQHmhtLawrwZgmAiTZ8Gzzz5bjjvuuCawTRAaCZgTpMZf/MVf9BWMZtmEwOvWret5nf/yX/5LufHGG5vHqh577LGmL92//bf/dlaC2Tx2Au1nnnlmwmUWLVo07uuW1zRzSfbzGkTWS0CdALuOCTCTFlKYnFAh4UJbJm067cwPlVNHQtyZqgxO6PC9554ud//p5rJ95Lyts+fnfCegYNgtpEB5Pr4fTGegvJB6JNtXAzBMDirMmoSi7WriBKEJmW+66aYm0J1OP/7xj5sgtS1h6v6qgodF5+uW1ypBck4rV65sqrT7kdchoTQA0yctLdpBcoLjiy//7EiYfHKZaQkzcjp75bnlvnvvaYKO2q8523jE0qXNfcDsO/roo8utt946GijXfsrzveXFfFGD3Roo10n5phoom2wPAGbPAYWhkV7FCUpTIdyWit2EzKnKPe2005p2EPuTYPRf/st/2azzK7/yK/uskxYbub9WSKeqObZs2dJcb4fMnWOlirczhM5tCcDrdma5jNm5XCqIs2zun2iZyUqQnBA529CW7crzr69DnkuV6uzaNiOvSfu553l3vu4ZC4D+vTAS1HZWpn3hyzfNSpDcKaFxQo3lxxw3ets377it2WZgONRAOecx3gR9DK/OX3wkAP7KdV8sU3H7zTcKkgFglgiTh0xnhWzCz7SjSE/gtKRI5e2nPvWp/QabCT4T3KbSOZXG/+gf/aNmnYTFkds/+9nPNuH17bff3oxddYa7GSvrpfo3Y6X9RW5rB95pS5FtTLVzlsspj19D6kj7igS2v/Ebv9E8ZsbKuO1lpirVyvU5RsLgPM88/zxenme2owbO2ZZaAZ7XJNtV23t0vu4ZO2N1C/IB2Fc+9NceyalInq5JmCarbtMRv2yzsWf37mabgeEhUJ7bOgPlrU89UaZi65NvrS9IBoCZpc3FEEkQmhA2oWYk2E2gmbC2hpypvk2VbQ06x5NgNMu110mIeu+99zaX084iYXDC5P21hMg6UUPkOla2K2Fru99yxsw2VT/5yU+aauEq63VO7Jflcz0BerZlquo2ZvtyOcF7XssE6dVf/dVfNc8rr1G7L3S2v/1aJDzOqd6W87x++feZ6kSHAAtJJtxrhwZrhyxIrhYvWdIEyv/6dy9prmebtz71eNPTGRgOWl7MbQl8s/9/9MG/bHrlT8UnP3VJuW/LnzfjfOSjqwsAMHOEybMola41cK0VwQmIa/iZNgztgLRKRW2u19C0U10/IW2WSXCcVg39ShVuQtbOx0iY2llR3Dl5XbaxXWWdMdrPJduUcLZu5yDC5E41CK6vQ33cyayb9XKajYkJAeayu/9s8+jls1aeO2OT7E1Gti1hR+3tnNYcV68TJsMwESjPbQl+BxH+5v3kLL3tAWBWCJNnUapyUz0baVtRK5GrGmSmb28/MlbGyboJaTuD3n6Mt25nBXAvsmwC6ASyeV5ZbyrbNdFjtLcvj5UK6lR853Wop16kAjmvY11XiAwwOY8++MDo5fZPnIdVqtxqmLz9uWea9hzDWEnNwnD99dfv08bhvPPOK+eff36ZjM2b3/xyZ7LrR8LbOOyww0q/Ev7mOWWMfHG/Zs3k9gkCZQCA2aNn8izKQXSqiHNK796ErO1eyDXE3Lt377in8QLOhJ+1T3B6GOeUdhSTCUPz+ONNkJcWFtHPmHWyvbS6yLandcdEbTomK5Xc7VYV6dGc55DHyuuQ83abjYlkO2t/5/oaPvroo/ttCQLAvtLiosqke8NclVwtP/b4csRR72kup3fyiztNxMfsSZCc44+EpvV0xhlnlMlK8JrTVGzcuLE5TUZ+kXf66ac35wmDH3744TJZeigDAMwOYfKQyAeFnHJwXdtD1HYW7UnlIhW3nbdVtVo4lcm1Cjfj9dreoS2Vw6nQ7ZwUMLf1G6wmyE1o3l5vvKB6shLCt/sZ17YWud4OvSd6zPZzrMvUSQq7rQvA+F54/q0g9shfBrRzwa+dePLo5e89+3SB2ZRK2wSw9ZTwNMFpZ6CbY8MaziZQTQVwjitrNXKnLN8OXrNuO9zNelk/49QAui6TU/vxcz3LjVdJ3bZjx47m+eQ51BB4KgTKAAAzT5g8RDJZXLs6uQbMqepN/+IaLKdqNtfHk+A0y2WMGjpn+c5AuL3cRMF0KocTpmb9LFMn3qsT2PUjrTqyXh0nj5uxJiNjZKycMs4555zTBL/ZptovuvZozuuU1yGPuX79+tFJBassk+eY5RJGR0L03FZfw/q8JxPIAyxkL7bC5OXH9NeyaTYdsuStthbt5wDDIi0mOit7E+RGgt8EqsuWLWvaWST0HS9QztwVE4XJGSvr5Tg04e+FF17YBLUZs54SbEeOy3KMleVyuuqqqyasOM44GTvbt2LFitExpmK8QDmPM1GIDgDA1OiZPEsSYr7rXe/a57YEogk1a5CbthAJMhOW1onq0qohbTGqBLV1rBz0Z4xUJtfK2lzvfLy6XCpOcl/aOHT2Bs71tMhImJ3ANhK0ZpvaFcbtx2+v2x4rQXmeRx0n69fbqozRrXVG7s+HlnYAnrGynZ3V0rktj3faaaeNec514r/Obcvrntchz7HeViu8U+Gck+pkgPlPj2SGydq1a8uhhx46ej1hbELb9BtOaJtANqFwJluu4eydd945Gq7W5frpk5x2GlH7IidYTkhdq4qj9ibOfTm+qo+ddXLbeEFx1sk4WabfwoT96eyhHAm4Yyr9oQEA2NeivWlgy5xQw+Re9TNB3nQ8/kRjxFTHmanHnI3tBRaWx767o8xXm761cXQyu9UfXzMnJuCLubrdk3HqB5YVhlcqeBMGp4q3SrCcMDYtI37rt36r+fI8wekHP/jB0eA0YW4C5K1btzbX232KI6Fr1sntmdAv2vclkE0gXSuMc33Dhg3N8u3lIl/ctye9S2Vwti+T4bXViukrr7yyGSPbmvVqO41ByGO0A+VIpfRkJ/qbafP5/YCpsa8GYJioTJ5D+g00BxkkT+bxp2uMmXxMITIAMJsSzI7XXzitJhLGJvB96KGHRsPdhLO5LSFq1qvXu6khcGSs/KIrgXBuSwg8kYTbWT7bsz8JpxPqZtyE41lnEG0u2sarUE4ld55bfX0AAJgaPZMBAGAOSjCbsDRBbg2c0/Ki3Y7ikUceGXfdhMAJoSPBa7uF2K5du5q+wwmSc1/nhHYJZ6tUGGfdOqlexhxvPo5sY709y2X8u+66a+BtKGqg3K6WzvXOCQsBAJgclckAADCkOquCU2Fbq2zToiJBabuNQ0La3J9q5DppXrvncudyaVORALZdJZzb87i5vbMyOuFvHb9WG6dlxurVq0cfp04G2JZtTShdl0tgXdttfOMb3xitih6EbG+qqvMc8ph5DTrn1gAAYHL0TAaAWaRn8vDRM5n5IBXFtb9yt+XGa6ORIDqB73j31fvbY3dbvt/lBiGPdcMNNzSh9Uw83lTpmcxE7KsBGCbaXAAA0+K0Mz9UFi9e0lxefuxxZa5YcdLJzXm2/axV5xaYi2pVcS/LjWeiXs3t+/tZvt/lBiGPtW7dujkRJAMAzBUqkwFgFs33SrQ9e3Y35zVUnivm6nb3S7UbDA+VyUzEvhqAYaJnMgAwbeZqGDvfQ2QAAIDJ0OYCAAAAAICuhMkAAAAAAHQlTAYAgDnm5ZdfLrfeems5//zzy4UXXliuv/765rZeff/73x8z1jnnnFO2bdtW+nHZZZc12wAAwMIhTAYAgDkkQXAC5ITACZGvueaasmvXribc7VXWefjhh5vLhx12WFm3bl05+uijSz+uuuqqJswGAGDhMAEfAADMIVu2bCmnn356E+ZWCYMTJm/cuLGsWbOmCYoTEud869atzfI1+N28eXPZsWNHc55AetWqVWMqlVNtnNvuvPPOsmLFima8LJux2uNkjGXLljUhdB63szI662UbMnbGyv3t9QEAmHtUJgMAwBySMHm8QPa8884brTbOeaqPE+Bm2XvvvbepYo4ExAl5c55TJOxNdXMkTM5jJFDOeaqgEwhnnNyXYDkyZm2NkUC5nrLsXXfdNRokX3rppU3onPUTOucEAMDcpDIZAADmkFoR3Cm3tauDTzjhhCbIrZdXr17dXM/lBL05n6i1Ra0qjoS/dZwEwu0q5irBc+S+LH/LLbc01xNSZ90afifgTkV1xgcAYO4RJgMAwByT0Lhbj+NadRwJhmvVcA2J96cu08uybQmOU8lcty2Vy6mSTqVyNV4YzfTY+tTjZc/u3eW0Mz9c5qJN39pY7rv3nrJ4yZJy9bpry+LFS8pM2/7s0+WF53fO2dcQAAZNmAwAAHNIbT+RyuK2BLbt21LBPJPSRiPb1q46ThidquT0Smbm3b/lz8tfjwTKgwhCH33wgbJnz+5y9spzy0y4b8s9ZdO3N5bVF6wpy485flaC5Hj0wb9sAm1hMgC8Sc9kAACYQ1L5m+C49i6OOkFe7qvS07hWASd8Tk/kdtjcOWHeVKS1RR6vPSlgrFy5stm22k85j9nebuaOx0ZC1YTTM2XbU0+UFSeeUlZ/fM1IkPuhAgAMB5XJAAAwhySU3bBhQxPc3nDDDc1tCYlzW7v1RaqB6yR8CZVrH+NIBfHatWubKuLOAHgyMjFfHueMM84YvS2hcaqS0+oi/ZoPPfTQfbaD4ZHK4+3PPtNcPmvVueXIo5aO3pcq4Ree/2F58fmdzeUjl75nJOg9ubkv1cr3j9yWdhppR3HqmR8es+5Etj75eNk6Ehj3+3jj6WUbet3Ozu3qZfuzbStOOqUAwEKwaO+IAgDMise+O7M/Q4e2Uz+wrDC31erizt7GCXcjPYwT4I7XX3midadDHiuV0d36PC9k0/F+cPvNNzVtLr70h7dNuEz6AX9lwxfLizt3NgFrAtz0Wk5FcFpMxHXrP9+Eu6+MBLHLjz2+qRT+yEdXN+v+/u9d0dx2xEgwm/UWjSy/dt21EwbKCXW/ct0Xm+A6AWzC3W6Pt/zY48onL7pkwu2/bv3nmmWz/RnrlZHHuPh3rhhtTTHRMld/4YvN+FV6NKe1RiqiEzhnuVPPeHP59mvYXi5yf16PibZxquyrARgmKpMBAGCO6iUInijAnYkQuf1YM/l4Ca9TEa1Xc3ebR4LRhKxf+PJNowFwgtIEpitOOrkJTK9e98UmmE4om8vV3d/ZVI5Y+p4xt31lw7XNpHUThckZN4F1+/Huv/ee8rWR8bs93nhuv/nGJsD+N1+5bbSv8h/fcVu5/as3jYbJCZLfPxJIX3z5Z8csc93vf778mz98c73aozkT/dUq6BqWH7LkrX7Ndbl1X7pxNIgeDdWPOa6cvWpmekoDwGzRMxkAAOaZhKgLNUhNe430jk4bj0H2hZ6vMrnc+ResGRP+pkr4iKPe00w+tz8JoUvZ24Sp1WfWfm6/k9Wl1USqeNuPd9bKc5sA974+ezK/0FRRPzEmJI5UCH/hSzc1lxNsJ7xe/fELxyyT55iq6Ef/5wPN9cce+ssmyG6308g2dj6XBN/Z3nZFc10uYwDAfKcyGQAA5pmFGCRnAsK098h5leuD6Ak9X9UQ+P3HHL/Pfb82Eqq+2AqJx5MQOlW///p3L2lC2ASq++uZnMdLm4t2EFvlttqzuVcv7vzhL9c9bp/76jbU59i5TNpYJDCvz3HP7l2jbSvGbtdxTauQt57D8806ed5t2fbxnhcAzDfCZAAAYM5KT+hMNNgOkSNtNZYt02t2shL67mkqjyeWwDa9hFPBnL7B//07m5r2Ed2qk2fS/p7D4sWLxyyX59zL+gmYTx3n+S1efGgBgPlOmAwAAMw5aWGRyuONGzeOuT0h8po1a5rTTPZpnovS9iGnBMGdlbvfe/bpngLhBLCZkC+ntJdIxe799/75uOsmfG4e78nHx7STiNzWb2Vv+jVHWlm0q4qzTdufebosP+74pldyM/5Tj49ZJhXL2597ZnTSv/c3ldFP7/MYWabtzfH2lrNX6o0MwMKkZzIAADBn1BB59erV+wTJ5513XrnzzjvLpZdeKkjuQVo9nLXq3GYivXbf4zpJXnobt+3Zs2vM9fW/d0Wz7Fv37x4zzngy5t1/Nvbx8vjtYLdXCacTSqcauj3eN7/+5gR8Ca4TUNdl2pXHecx2T+SzVv3dpv9yekhXCZcfffCBju0/v6nEbt+ex/5Xv3tJMyYAzHcqkwEAgDkh4XGC5M6J9dIjOgHyQp10cH8SCl/yidX73H71umubkDUT0aUquPY9bqp6n32mqTIeM0neSNj66HUPlP/n4jVNIJzg9+/9X6vL126+qTw2EqweMbJs1ktA/YmRdSeSdVMl3Pl4qzsmAexVJt9LNfTvjwTbqa6ufZevXvfFcZep29m5TKqWsw23jzyfTBKY7coEg3l92gFzXe4rG64dra7OeEcsXTpu6wsAmG8W7R1RAIBZ8dh3d5SFLiFGfqo8mRCBqTn1A/rJMjds3ry5CZHTH7nthBNOaCbYmw8h8nS8H6Syds+ePePel+A1lbtVqm2zfK1Wbt9XpQI3QfD7jzlutCVFbkuYnN7CCVRT6Tveup1SBZz9/0SP92bLiUXjTq43nrr9+9uGXpbp3K48r4TKndvRft4rTjpln7Ydg2RfDcAwESYDwCyajvAgVVXtKqrIh+Dlxxw/6cqv6ZIP4f/PP1/TfAhPldxUjFd5V932zeH76fELO3eWxYcu6Sl0mS4CCoZdJtVLiDze5HpXXnllOf/888t84ctFJmJfDcAw0eYCAOahVFR9Zu3nR6+nEuu/f2dT2bD+8bJ2JLQdlkA523nx5Vc0VW6DkAmRUkk2F9z+1RtHfy4NjLW/ENnkegAAs0eYDADz0CGLl4z5yW0unzUStKZn5O033zhuFXAqZePIpUvHva9W0Y633P7WjfSe3LNr97iVuCt+/ZTm9vZYdZxu43Y64peTMU1kvGrgum3tx9zfc53ouXV73RLoH7Lk0CbIr+M2k1XtZ5tmu3IZZpoQGQBguAmTAWCBSBXwJz91yUig/Pmm52WqYiN9HxMwp09klQmH2hWzqaI94qj3lBef/+HocgltM0HTH99x2+htCUq/8OWbRgPQ8cZOqJ3tqMt0Vuj+6//7kqZaedO3NjYTR9VxB1FRnceKdpie7cvkSf/mD28bffw8r0cffGDM8/rM2s+N9gjt9XW77vc/X85e+XebtiN5Lrk/ldN5jMi6d39nU7n4d64oZ4/c3suYMB8JkQEA5gZhMgAsIAlt36ySfWY0TE61coLmhKkJTRN8fvPrtzW3feSjb/UhzkRDCTU/c/Xnm8mJ0ps5YeknRoLXhLMZM2Pd/aebRsPPXF9U3gxvEz7XZfI4+wtIE7D+88s/O2ad3JaQdyouHhnz93/viiYozuRL9225p5mQqQbJ1aZvb2xC7zzX9HXePBJs57lmuRqC19dt3ZduanpST/S63f1nm5rrGS/3Zf2M85XrvthM2pT7cnts/mWAXv8tsp1f2XBts629TkIFc4kQGQBgbhEmA8ACk1YLCUgjYWrCy6vX3TZa9Zu+wy/u3NkEqu1QNFW59XrCzftP+vNmnCz/5v3HNbdvf+6Z0XUSItcAdaJlxpMxa7uKrJNq3gSr3cLk+0cC3VRdt2W763p5jnkO37zjtibIzXMcb1LCBO11nWz7J35ZqVyD8lye6HVLb+r263bqGR9uqovbsk59XdqP/ddPPdG8PvW2XK7BMswnE4XIcd5555XLLrusHH300QUAgOEiTAaABWzbSHiZ4LQzrDztzA81QWvaLtT7juhYJkHoIeP0861BdWTd9P+t4WvG++uRsPfIo95T9qdz3F77Br/ZM/mUsbd19DJO0Jsq4lo13Rn0Rmff5QS/7RA8bSjy3FKhvbW1XJ5rnmfO6zb3EwSnJUZe97yG+TdIEC5IZr54+eWXy1133VXuvPPO8v3vf3/MfSqRAQDmBmEyACwgCSnT97gdsB4xTliZ6uV4sRUmT0b6Hqc9RcZ7/7HHNY81neFou/fyRN4Mhj80sl2bxw2S45AJwusalL8ycp7L99/75+Nuw2Rl2/Nvc/+WP2/aiCSU7uwxDXNNQuSNGzc2p1xuEyIDAMwtwmQAWEBSIRw18DxkJFjtbAsRL+78YXN+xBSC31Qhp8o27SLabR8Skr7wy4n1ZkMeO0FyXoP0M04Ljc6Ae7w2HC+MhPCLfxmy53XJa3f1ui+WQUu7jNo6JP9eeb269ZiGYZTq482bNwuRAQDmkQMKALAgpCVDegUn2H2rJ++HmgnuOgPlVNweOcUq4ld272rOTz3zw2Nu/+txwuuZdPvNNzZBcoLg5ccc11zvlMkGUxVcJYBOa4tTR16vWHHSyU3Vdufrlh7UOfWq/RipdP7jkX+f9ph14r32cjDs0gc5PY/PP//8pi9yO0hOcHzppZeWTZs2NeeCZACAuUVlMgDMQ6+MhI9fu/mm0evbn3u6CY3PXnXumEnsEqqmCjbVrwmZ097hsYceGAmen5hy1W36/aaCd8P6z5Xzf9lOIhPkpUVEOapMi0dHtn28qufaJiJBb0LhTGoXF1/+2fL7v3dF04qjXT2dquPcXm/LpHoJdWvFcG2n8ZUN1zYT7KXHcqqZM06vFcS1MjrSKzmv1/Znn26C7PpvkZ7W2d5PXvTpAsOs9kNOJfK2bdv2uf/0009vJtZLwAwAwNwlTAaAeeaEk04ueztuS3CZQLJzYrm4+PIrmgnp0qc3QWyWaap2R8LNKhWyCVg7H2e8xz5i51th7tXrri2bm77Jm0cnsUtbiRdby+S2djuN9Ag+smPSvAS5p+35cNmfs34Z9O5PqokTpteK65x/YiRoTtDe9pF/OBIiLyrl7j/d3FxPCN8OmyP9lrPded02ffuPm57QeS3Pbm3Hm5PoHTfutmS8VBznsV84cWfzen9m7eebQPrRB/+y+bfImHkNJxqDt6Qa9vrrry+7du0qt9xySzn66KML0y+v+5YtW5ogubOVRSRETgVyzgEAmPsW7R1RAIBZ8dh3dxSGyyWfWF0u/p0rmgB5vjv1A8vKfJAQOX15q4SXOTE9ahVyQuSEyZ1qP+RUIQv1e+f9gInMl301APODymQAAOakTPB21VVXjWmrkCBTK4XpkeA4PZDzek9Uhbxq1aqmnYVeyAAA85MwGQCg5Yij3tO05GC4pRK5c3K3hJnXXHONatgB6tbGIqFxwuOEyFpZAADMf8JkAICWL/3hHxWGVwLN9evXNwFnlUAzbS3SWoGp6xYghypkAICFSZgMAMCckJAzlcdpb1EtW7asbNiwoZxwwgmFyUlgnNYV3QLk2gs5IbLXGwBgYRImAwAw9Don2YsEm6lIVhnbvwTydQK9nPYXIGtjAQBAJUwGAGBoTTTJXqqRhZu9S1hcg+OEyO3q7k4CZAAAJiJMBgBgKCXwvPDCC/epmk3QGbldVfL48to99NBDTQifALkdxo8nkxYmPBYgAwCwP8JkAACG0ubNm8dtv5B2F7XlRcLk9O+t5+3LCyVoTnC8devWJjTO5f21rajqa1QD5ITJAADQzaK9IwoAMCse++6OArPl1A8sK8NsosrkXiUwTUhag9NM1lev18tzRV6DXbt2NaFxDY9TbZzLvb4+qTiuAfJCCtvnCu8HTGTY99UALCzCZACYRcIDZtNcCCjS33ft2rWj12sonAA1Yepkg+YqgXLGa19OyHrooYeO3t6+f9BqSJzzHTt2jF6vl+tz3F+P4/HUAD0Bcg2RhcfDzfsBExEmAzBMtLkAAGBopYp2zZo1o20tEqpef/31o8FuDVwTvua+9uVeguYsN5mgNmFzVcPmbrJdVQ2QB6Hd4iOnM844Q9sKAACmhcpkAJhFKtGYTXOl2i2ha9pd1NA3lba33HJLT+u1K3vbIXO9PJckIF6xYsVoBXW7RzRzn/cDJqIyGYBhIkwGgFkkPGA2zaWAIqFwAuXqqquuaiqWp6qGzNEOm9utJer90xE+1yrnnLd7PCcszu0Jj+vtzG/eD5iIMBmAYSJMBoBZJDxgNs21gCKtLtLiokp1cqqUZ1rta1wv99quot0OowbIUHk/YCLCZACGiTAZAGaR8IDZNBcDissuu6w8/PDDzeWEs9/4xjeEsswL3g+YiDAZgGFyQAEAgDnimmuuGQ2P037i1ltvLQAAwMwQJgMAMGekd/C6detGr6f1xebNmwsAADD9hMkAAMwpq1atGjP53g033NBz32IAAGDyhMkAAMw5l156aVOlHAmSax9lAABg+giTAWAWHXiAt2Jmx8EHHVjmsvRNvv7668sJJ5xQTj/99KZaGeayg982t/9PMj3m+r4agPln0d4RBQCYFT/40cvlBz/283xm3lHvOrT86pHvLMBw8H7AeN77K4eV9777sAIAw0I5FADMoqMOX1IOeftBBWbSoYccLJyAIZP/k+9a8o4CVf4e7KsBGDYqkwFgCPzo5T1l1ys/KzDdDj3k7eXdhy0uwHD6ye5Xy65Xf15ef/31wsJ04IEHlnctfnuzvwaAYSNMBgAAAACgK20uAAAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANCVMBkAAAAAgK6EyQAAAAAAdCVMBgAAAACgK2EyAAAAAABdCZMBAAAAAOhKmAwAAAAAQFfCZAAAAAAAuhImAwAAAADQlTAZAAAAAICuhMkAAAAAAHQlTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTIAAAAAAF0JkwEAAAAA6EqYDAAAAABAV8JkAAAAAAC6EiYDAAAAANCVMBkAAAAAgK6EyQAAAAAAdCVMBgAAAACgK2EyAAAAAABdCZMBAAAAAOhKmAwAAAAAQFfCZAAAAAAAuhImAwAAAADQlTAZAAAAAICuhMkAAAAAAHQlTAYAAAAAoCthMgAAAAAAXQmTAQAAAADoSpgMAAAAAEBXwmQAAAAAALoSJgMAAAAA0JUwGQAAAACAroTJAAAAAAB0JUwGAAAAAKArYTL8/9uxAwEAAAAAQf7WIyxQGAEAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALBkMgAAAAAASyYDAAAAALACPohiw1T/6hcAAAAASUVORK5CYII="
-    }
-   },
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Why We Need Feature Engineering\n",
-    "\n",
-    "The next question is: why do we need feature engineering in the first place? Isn't machine learning supposed to magically extract the information it needs from the data? Unfortunately, it is not quite that simple. Standard machine learning algorithms require the data to be presented in a *flat table*. \n",
-    "\n",
-    "We say *standard machine learning* and that means that there are exceptions. *Deep learning* can automatically extract features from unstructured data like images, text and speech. On the other hand, *relational learning* can automatically extract features from structured relational data.\n",
-    "\n",
-    "In the absence of these two approaches, we have to manually engineer features. This involves interactions between *data scientists*, who understand statistics, and *domain experts*, who understand the underlying problem domain. For example, domain experts provide domain know-how to data scientists, who write code and software that needs to be evaluated and optimized by both parties. The result is a flat table containing features to be used in machine learning. However, the process of getting to that table is cumbersome, time consuming and error prone.\n",
-    "\n",
-    "![Manual Feature Engineering](attachment:ManualFeatureEngineering2.png)\n",
-    "\n",
-    "In this figure, we refer to *relational data*. Relational data is structured data that you would usually find in relational databases such as PostgreSQL, MySQL, MariaDB, BigQuery or Redshift. However, the same holds true for time series. In fact, time series can be considered a special case of relational data. Below, we will see why that is."
-   ]
-  },
-  {
-   "attachments": {
-    "FeatureEngineeringExample2.png": {
-     "image/png": "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"
-    }
-   },
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Interesting Features for Time Series\n",
-    "\n",
-    "Two important characteristics are *trend* and *seasonality*.\n",
-    "\n",
-    "- The *trend* indicates whether your time series increases or decreases over time.\n",
-    "\n",
-    "- The *seasonality* indicates any patterns that vary by hour, day etc. For instance, the Interstate 94 dataset has very obvious seasonal patterns: Traffic volume is highest in the morning (when people drive to work) and in the evening (when people get home from work), very low at night and overall lower on weekends.\n",
-    "\n",
-    "But there are many other important characteristics, such as the *average*, *median*, *max*, *min* and *standard deviation* of your observations over time. So when you build features for time series, a very common approach is to use a *sliding window* technique. Whenever you want to make a prediction you take all the values over a fixed period of time right up to when you want to make the prediction and then apply all sorts of aggregations to this. For classification tasks, every window or time point is classified by a certain label.\n",
-    "\n",
-    "![Feature Engineering Example 2](attachment:FeatureEngineeringExample2.png)\n",
-    "\n",
-    "This depiction shows how we can engineer a very simple feature for a time series. As we would for relational data, we define some kind of criterion over which we identify the data we are interested in (in this case, the last five days), which we then aggregate using some aggregation function (in this case, the average). This is a very simple example how feature engineering for time series usually works. But if you engineer features for time series in this way, you are effectively thinking of time series as relational data: You are identifying relevant data from your data set and aggregating it, just like you would for relational data. In fact, what we are doing is effectively a *self join*, because we are joining a table to itself.\n",
-    "\n",
-    "This is just a simple example. But there are many more features you can generate. For instance, for the EEG data, you build features like this:\n",
-    "\n",
-    "- Average EEG value in the last second\n",
-    "- Maximum EEG value in the last second\n",
-    "- Minimum EEG value in the last second\n",
-    "- Median EEG value in the last second\n",
-    "- First measurement of EEG value in the last second\n",
-    "- Last measurement of EEG value in the last second\n",
-    "- Variance of the EEG value in the last second\n",
-    "- Standard deviation of the EEG value in the last second\n",
-    "- Exponentially weighted moving average of the EEG value in the last second\n",
-    "- 1%-quantile of the EEG value in the last second\n",
-    "- 5%-quantile of the EEG value in the last second\n",
-    "- ...\n",
-    "\n",
-    "Of course, we mustn't just assume that one second is the right period of time to use. So we could take all of these features and calculate them for the last minute or last three hours.\n",
-    "\n",
-    "Moreover, the features we have discussed so far don't really take seasonality into account (with the exception of first measurement of EEG value in the last second, because this is the measurement from exactly one second ago).\n",
-    "\n",
-    "On the other hand, the Interstate 94 dataset is strongly seasonal, we should take that into account as well. So we could calculate features like this:\n",
-    "\n",
-    "- Average traffic volume in the last four weeks, but only where weekday equals the weekday the point in time we want to predict.\n",
-    "- Average traffic volume in the last four weeks, but only where hour of the day equals the hour of the point in time we want to predict.\n",
-    "- Average traffic volume in the last four weeks, but only where both the weekday and the hour of the day equal the point in time we want to predict.\n",
-    "- Maximum traffic volume in the last four weeks, but only where weekday equals the weekday the point in time we want to predict.\n",
-    "- ...\n",
-    "\n",
-    "What should be very obvious at this point is that there are *many* features that you can generate like this, even for a very simple time series problem. When you have a multivariate time series (meaning you have more than one input variable), you can apply these techniques to every single column in your input data and you will get many, many features. You would very likely have to apply some kind of feature selection techniques to focus on the most useful features.\n",
-    "\n",
-    "The beauty of this approach is that it is very flexible and uses few assumptions. We have noted above that many time series are *not* equally-spaced. This would be a problem for classical time series analyses like ARIMA or ARMA. Not here. Nothing about this approach makes any assumptions about the spacing. "
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Automated Feature Engineering for Time Series\n",
-    "\n",
-    "You will find a lot of examples where people conduct this work manually (mainly using pandas). But today, we are not limited to manual feature engineering. There are numerous tools and libraries which can automate away this kind of work.\n",
-    "\n",
-    "Unfortunately, automated feature engineering has been getting a bit of a bad rap, mainly for taking too long. And it isn't wrong: Some of the more well-known libraries like featuretools or tsfresh are slow and not very memory-efficient.\n",
-    "\n",
-    "Overall, the features extracted from time series by such libraries are quite similar [[Henderson & Fulcher](https://ieeexplore.ieee.org/document/9679937)]. However, the stark differences in terms of runtime and memory consumption make it worthwhile taking a closer look as some of the newer tools and libraries like getML or tsflex are highly optimized and can generate many features in a short period of time. \n",
-    "\n",
-    "In the following, we would like to introduce time series classification using the getML Pyhon API. Above, we have discussed how time series can be seen as a form of relational data and introduced the term relational learning. We can utilize a very simple relational learning approach by interpreting the time series as a form of relational data and conduct a self join."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### An Introduction to Univariate Time Series Classification with getML\n",
-    "\n",
-    "In this tutorial, you will learn how to use getML to classify univariate time series. We first explore the data and learn what we are dealing with. Subsequently, we demonstrate how to efficiently build a full fledged machine learning data model and how to use getML's automatic feature learning algorithm FastProp.\n",
-    "\n",
-    "### About the Dataset\n",
-    "The original dataset from the reference comprises 500 files, each file representing a single person/subject. Each recording contains the EEG signal value of brain activity for 23.6s sampled into 4096 data points. These recordings have been split into 1s windows. This results in 23 x 500 = 11500 windows of EEG data over time in 178 datapoints and each window is categorized into 5 labels:\n",
-    "\n",
-    "1. Seizure activity\n",
-    "2. EEG recorded at tumor site\n",
-    "3. EEG recorded in healthy brain area\n",
-    "4. eyes closed during recording\n",
-    "5. eyes open during recording\n",
-    "\n",
-    "Subjects labeled with classes 2-5 did not have epileptic seizures. We can thus do a binary classification of subjects suffering an epileptic seizure or not, meaning classes 1 or 0, respectively.\n",
-    "\n",
-    "### Acknowledgements\n",
-    "Andrzejak RG, Lehnertz K, Rieke C, Mormann F, David P, Elger CE (2001) Indications of nonlinear deterministic and finite dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state, Phys. Rev. E, 64, 061907"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Setup\n",
-    "\n",
-    "### Start up getML\n",
-    "\n",
-    "First, we import the necessary libraries  and launch the [getML engine](https://docs.getml.com/latest/user_guide/getml_suite/engine.html). The engine runs in the background and takes care of all the heavy lifting for you. This includes things like our powerful database engine and efficient algorithms as well as the [getML monitor](https://docs.getml.com/latest/user_guide/getml_suite/monitor.html), which you can access by pointing your browser to http://localhost:1709/#/"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
-      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
-      "Note: you may need to restart the kernel to use updated packages.\n"
-     ]
-    }
-   ],
-   "source": [
-    "%pip install -q \"getml==1.4.0\" \"numpy<2.0.0\" \"matplotlib~=3.9\" \"seaborn~=0.13\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import numpy as np\n",
-    "import pandas as pd\n",
-    "\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "\n",
-    "sns.set_style(\"whitegrid\")\n",
-    "\n",
-    "import getml"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "All your work is organized into [projects](https://docs.getml.com/latest/user_guide/project_management/project_management.html). You can easily set any name for your current project. The engine will create a new project or use an existing one if the project name already exists. It will also provide you with a direct link to the project within the monitor."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Launching ./getML --allow-push-notifications=true --allow-remote-ips=false --home-directory=/home/alex/.local/share/hatch/env/virtual/getml-demo/txflr3_Z/getml-demo/lib/python3.10/site-packages/getml --in-memory=true --install=false --launch-browser=true --log=false in /home/alex/.local/share/hatch/env/virtual/getml-demo/txflr3_Z/getml-demo/lib/python3.10/site-packages/getml/.getML/getml-1.4.0-x64-community-edition-linux...\n",
-      "Launched the getML engine. The log output will be stored in /home/alex/.getML/logs/20240826151418.log.\n",
-      "Loading pipelines... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n",
-      "Connected to project 'epilepsy_recognition'\n"
-     ]
-    }
-   ],
-   "source": [
-    "getml.engine.launch()\n",
-    "getml.engine.set_project(\"epilepsy_recognition\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "You can manage your projects conveniently through the monitor web interface or directly using Python commands. For example, you can suspend your current project to free resources using [`getml.project.suspend()`](https://docs.getml.com/latest/api/project/getml.project.suspend.html), switch to another project on the fly using [`getml.project.switch('new project name')`](https://docs.getml.com/latest/api/project/getml.project.switch.html), or restart using [`getml.project.restart()`](https://docs.getml.com/latest/api/project/getml.project.restart.html) should something go wrong. You can even save your current project to disk using [`getml.project.save('filename')`](https://docs.getml.com/latest/api/project/getml.project.save.html) and load it with [`getml.project.load('filename')`](https://docs.getml.com/latest/api/project/getml.project.load.html)."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Load Data\n",
-    "\n",
-    "The original dataset was hosted on the [UCI repository](https://archive.ics.uci.edu/ml/datasets/Epileptic+Seizure+Recognition) but was unfortunately removed. You can get the dataset via Kaggle, [here](https://www.kaggle.com/datasets/harunshimanto/epileptic-seizure-recognition).\n",
-    "\n",
-    "The dataset we will be working on is stored in a CSV file located on disk. As we will perform data exploration, we will first load the data into a pandas DataFrame, as usual, and examine the raw data."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>Unnamed</th>\n",
-       "      <th>X1</th>\n",
-       "      <th>X2</th>\n",
-       "      <th>X3</th>\n",
-       "      <th>X4</th>\n",
-       "      <th>X5</th>\n",
-       "      <th>X6</th>\n",
-       "      <th>X7</th>\n",
-       "      <th>X8</th>\n",
-       "      <th>X9</th>\n",
-       "      <th>...</th>\n",
-       "      <th>X170</th>\n",
-       "      <th>X171</th>\n",
-       "      <th>X172</th>\n",
-       "      <th>X173</th>\n",
-       "      <th>X174</th>\n",
-       "      <th>X175</th>\n",
-       "      <th>X176</th>\n",
-       "      <th>X177</th>\n",
-       "      <th>X178</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>X21.V1.791</td>\n",
-       "      <td>135</td>\n",
-       "      <td>190</td>\n",
-       "      <td>229</td>\n",
-       "      <td>223</td>\n",
-       "      <td>192</td>\n",
-       "      <td>125</td>\n",
-       "      <td>55</td>\n",
-       "      <td>-9</td>\n",
-       "      <td>-33</td>\n",
-       "      <td>...</td>\n",
-       "      <td>-17</td>\n",
-       "      <td>-15</td>\n",
-       "      <td>-31</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-103</td>\n",
-       "      <td>-127</td>\n",
-       "      <td>-116</td>\n",
-       "      <td>-83</td>\n",
-       "      <td>-51</td>\n",
-       "      <td>4</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>X15.V1.924</td>\n",
-       "      <td>386</td>\n",
-       "      <td>382</td>\n",
-       "      <td>356</td>\n",
-       "      <td>331</td>\n",
-       "      <td>320</td>\n",
-       "      <td>315</td>\n",
-       "      <td>307</td>\n",
-       "      <td>272</td>\n",
-       "      <td>244</td>\n",
-       "      <td>...</td>\n",
-       "      <td>164</td>\n",
-       "      <td>150</td>\n",
-       "      <td>146</td>\n",
-       "      <td>152</td>\n",
-       "      <td>157</td>\n",
-       "      <td>156</td>\n",
-       "      <td>154</td>\n",
-       "      <td>143</td>\n",
-       "      <td>129</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>X8.V1.1</td>\n",
-       "      <td>-32</td>\n",
-       "      <td>-39</td>\n",
-       "      <td>-47</td>\n",
-       "      <td>-37</td>\n",
-       "      <td>-32</td>\n",
-       "      <td>-36</td>\n",
-       "      <td>-57</td>\n",
-       "      <td>-73</td>\n",
-       "      <td>-85</td>\n",
-       "      <td>...</td>\n",
-       "      <td>57</td>\n",
-       "      <td>64</td>\n",
-       "      <td>48</td>\n",
-       "      <td>19</td>\n",
-       "      <td>-12</td>\n",
-       "      <td>-30</td>\n",
-       "      <td>-35</td>\n",
-       "      <td>-35</td>\n",
-       "      <td>-36</td>\n",
-       "      <td>5</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>X16.V1.60</td>\n",
-       "      <td>-105</td>\n",
-       "      <td>-101</td>\n",
-       "      <td>-96</td>\n",
-       "      <td>-92</td>\n",
-       "      <td>-89</td>\n",
-       "      <td>-95</td>\n",
-       "      <td>-102</td>\n",
-       "      <td>-100</td>\n",
-       "      <td>-87</td>\n",
-       "      <td>...</td>\n",
-       "      <td>-82</td>\n",
-       "      <td>-81</td>\n",
-       "      <td>-80</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-85</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-72</td>\n",
-       "      <td>-69</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>5</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>X20.V1.54</td>\n",
-       "      <td>-9</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>-98</td>\n",
-       "      <td>-102</td>\n",
-       "      <td>-78</td>\n",
-       "      <td>-48</td>\n",
-       "      <td>-16</td>\n",
-       "      <td>0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>...</td>\n",
-       "      <td>4</td>\n",
-       "      <td>2</td>\n",
-       "      <td>-12</td>\n",
-       "      <td>-32</td>\n",
-       "      <td>-41</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>-83</td>\n",
-       "      <td>-89</td>\n",
-       "      <td>-73</td>\n",
-       "      <td>5</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>5 rows × 180 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "      Unnamed   X1   X2   X3   X4   X5   X6   X7   X8   X9  ...  X170  X171  \\\n",
-       "0  X21.V1.791  135  190  229  223  192  125   55   -9  -33  ...   -17   -15   \n",
-       "1  X15.V1.924  386  382  356  331  320  315  307  272  244  ...   164   150   \n",
-       "2     X8.V1.1  -32  -39  -47  -37  -32  -36  -57  -73  -85  ...    57    64   \n",
-       "3   X16.V1.60 -105 -101  -96  -92  -89  -95 -102 -100  -87  ...   -82   -81   \n",
-       "4   X20.V1.54   -9  -65  -98 -102  -78  -48  -16    0  -21  ...     4     2   \n",
-       "\n",
-       "   X172  X173  X174  X175  X176  X177  X178  y  \n",
-       "0   -31   -77  -103  -127  -116   -83   -51  4  \n",
-       "1   146   152   157   156   154   143   129  1  \n",
-       "2    48    19   -12   -30   -35   -35   -36  5  \n",
-       "3   -80   -77   -85   -77   -72   -69   -65  5  \n",
-       "4   -12   -32   -41   -65   -83   -89   -73  5  \n",
-       "\n",
-       "[5 rows x 180 columns]"
-      ]
-     },
-     "execution_count": 4,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data = pd.read_csv(\"data/Epileptic Seizure Recognition.csv\")\n",
-    "\n",
-    "# view first 5 rows of the data\n",
-    "data.head()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The first column contains extraneous metadata and can be dropped. The last column is named `y` and contains the class labels. As described above, we will do a binary classification into epileptic seizure (label 1) or not (label 2-5). Thus, we can set the labels 2-5 to 0, representing a non-epileptic instance, and 1 for epileptic seizure."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data.drop(\"Unnamed\", axis=1, inplace=True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# classify having epileptic seizure or not\n",
-    "class_relabeling = {1: 1, 2: 0, 3: 0, 4: 0, 5: 0}\n",
-    "data.replace({\"y\": class_relabeling}, inplace=True)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now we can check which values we have in the label column and the DataFrame in general."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Number of records epileptic 2300 vs non-epileptic 9200\n"
-     ]
-    }
-   ],
-   "source": [
-    "counts = data[\"y\"].value_counts()\n",
-    "print(f\"Number of records epileptic {counts[1]} vs non-epileptic {counts[0]}\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>X1</th>\n",
-       "      <th>X2</th>\n",
-       "      <th>X3</th>\n",
-       "      <th>X4</th>\n",
-       "      <th>X5</th>\n",
-       "      <th>X6</th>\n",
-       "      <th>X7</th>\n",
-       "      <th>X8</th>\n",
-       "      <th>X9</th>\n",
-       "      <th>X10</th>\n",
-       "      <th>...</th>\n",
-       "      <th>X170</th>\n",
-       "      <th>X171</th>\n",
-       "      <th>X172</th>\n",
-       "      <th>X173</th>\n",
-       "      <th>X174</th>\n",
-       "      <th>X175</th>\n",
-       "      <th>X176</th>\n",
-       "      <th>X177</th>\n",
-       "      <th>X178</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>135</td>\n",
-       "      <td>190</td>\n",
-       "      <td>229</td>\n",
-       "      <td>223</td>\n",
-       "      <td>192</td>\n",
-       "      <td>125</td>\n",
-       "      <td>55</td>\n",
-       "      <td>-9</td>\n",
-       "      <td>-33</td>\n",
-       "      <td>-38</td>\n",
-       "      <td>...</td>\n",
-       "      <td>-17</td>\n",
-       "      <td>-15</td>\n",
-       "      <td>-31</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-103</td>\n",
-       "      <td>-127</td>\n",
-       "      <td>-116</td>\n",
-       "      <td>-83</td>\n",
-       "      <td>-51</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>386</td>\n",
-       "      <td>382</td>\n",
-       "      <td>356</td>\n",
-       "      <td>331</td>\n",
-       "      <td>320</td>\n",
-       "      <td>315</td>\n",
-       "      <td>307</td>\n",
-       "      <td>272</td>\n",
-       "      <td>244</td>\n",
-       "      <td>232</td>\n",
-       "      <td>...</td>\n",
-       "      <td>164</td>\n",
-       "      <td>150</td>\n",
-       "      <td>146</td>\n",
-       "      <td>152</td>\n",
-       "      <td>157</td>\n",
-       "      <td>156</td>\n",
-       "      <td>154</td>\n",
-       "      <td>143</td>\n",
-       "      <td>129</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>-32</td>\n",
-       "      <td>-39</td>\n",
-       "      <td>-47</td>\n",
-       "      <td>-37</td>\n",
-       "      <td>-32</td>\n",
-       "      <td>-36</td>\n",
-       "      <td>-57</td>\n",
-       "      <td>-73</td>\n",
-       "      <td>-85</td>\n",
-       "      <td>-94</td>\n",
-       "      <td>...</td>\n",
-       "      <td>57</td>\n",
-       "      <td>64</td>\n",
-       "      <td>48</td>\n",
-       "      <td>19</td>\n",
-       "      <td>-12</td>\n",
-       "      <td>-30</td>\n",
-       "      <td>-35</td>\n",
-       "      <td>-35</td>\n",
-       "      <td>-36</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>-105</td>\n",
-       "      <td>-101</td>\n",
-       "      <td>-96</td>\n",
-       "      <td>-92</td>\n",
-       "      <td>-89</td>\n",
-       "      <td>-95</td>\n",
-       "      <td>-102</td>\n",
-       "      <td>-100</td>\n",
-       "      <td>-87</td>\n",
-       "      <td>-79</td>\n",
-       "      <td>...</td>\n",
-       "      <td>-82</td>\n",
-       "      <td>-81</td>\n",
-       "      <td>-80</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-85</td>\n",
-       "      <td>-77</td>\n",
-       "      <td>-72</td>\n",
-       "      <td>-69</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>-9</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>-98</td>\n",
-       "      <td>-102</td>\n",
-       "      <td>-78</td>\n",
-       "      <td>-48</td>\n",
-       "      <td>-16</td>\n",
-       "      <td>0</td>\n",
-       "      <td>-21</td>\n",
-       "      <td>-59</td>\n",
-       "      <td>...</td>\n",
-       "      <td>4</td>\n",
-       "      <td>2</td>\n",
-       "      <td>-12</td>\n",
-       "      <td>-32</td>\n",
-       "      <td>-41</td>\n",
-       "      <td>-65</td>\n",
-       "      <td>-83</td>\n",
-       "      <td>-89</td>\n",
-       "      <td>-73</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>5 rows × 179 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "    X1   X2   X3   X4   X5   X6   X7   X8   X9  X10  ...  X170  X171  X172  \\\n",
-       "0  135  190  229  223  192  125   55   -9  -33  -38  ...   -17   -15   -31   \n",
-       "1  386  382  356  331  320  315  307  272  244  232  ...   164   150   146   \n",
-       "2  -32  -39  -47  -37  -32  -36  -57  -73  -85  -94  ...    57    64    48   \n",
-       "3 -105 -101  -96  -92  -89  -95 -102 -100  -87  -79  ...   -82   -81   -80   \n",
-       "4   -9  -65  -98 -102  -78  -48  -16    0  -21  -59  ...     4     2   -12   \n",
-       "\n",
-       "   X173  X174  X175  X176  X177  X178  y  \n",
-       "0   -77  -103  -127  -116   -83   -51  0  \n",
-       "1   152   157   156   154   143   129  1  \n",
-       "2    19   -12   -30   -35   -35   -36  0  \n",
-       "3   -77   -85   -77   -72   -69   -65  0  \n",
-       "4   -32   -41   -65   -83   -89   -73  0  \n",
-       "\n",
-       "[5 rows x 179 columns]"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data.head()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Explore Data\n",
-    "\n",
-    "We first have a look at some common statistics of our data divided into both classes."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>count</th>\n",
-       "      <th>mean</th>\n",
-       "      <th>std</th>\n",
-       "      <th>min</th>\n",
-       "      <th>25%</th>\n",
-       "      <th>50%</th>\n",
-       "      <th>75%</th>\n",
-       "      <th>max</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>X1</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-8.992609</td>\n",
-       "      <td>70.455286</td>\n",
-       "      <td>-566.0</td>\n",
-       "      <td>-44.0</td>\n",
-       "      <td>-7.0</td>\n",
-       "      <td>26.0</td>\n",
-       "      <td>1726.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X2</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-8.877174</td>\n",
-       "      <td>70.560110</td>\n",
-       "      <td>-609.0</td>\n",
-       "      <td>-44.0</td>\n",
-       "      <td>-7.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>1713.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X3</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-8.910435</td>\n",
-       "      <td>70.372582</td>\n",
-       "      <td>-594.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-7.0</td>\n",
-       "      <td>28.0</td>\n",
-       "      <td>1697.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X4</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-8.969783</td>\n",
-       "      <td>70.030409</td>\n",
-       "      <td>-549.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-8.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>1612.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X5</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-9.085326</td>\n",
-       "      <td>69.377958</td>\n",
-       "      <td>-603.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-8.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>1437.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X175</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-9.848587</td>\n",
-       "      <td>69.550894</td>\n",
-       "      <td>-570.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-9.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>1958.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X176</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-9.620435</td>\n",
-       "      <td>70.353607</td>\n",
-       "      <td>-594.0</td>\n",
-       "      <td>-46.0</td>\n",
-       "      <td>-8.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>2047.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X177</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-9.395435</td>\n",
-       "      <td>70.934300</td>\n",
-       "      <td>-563.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-9.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>2047.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X178</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>-9.240435</td>\n",
-       "      <td>71.185850</td>\n",
-       "      <td>-559.0</td>\n",
-       "      <td>-45.0</td>\n",
-       "      <td>-8.0</td>\n",
-       "      <td>27.0</td>\n",
-       "      <td>1915.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>y</th>\n",
-       "      <td>9200.0</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>179 rows × 8 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       count      mean        std    min   25%  50%   75%     max\n",
-       "X1    9200.0 -8.992609  70.455286 -566.0 -44.0 -7.0  26.0  1726.0\n",
-       "X2    9200.0 -8.877174  70.560110 -609.0 -44.0 -7.0  27.0  1713.0\n",
-       "X3    9200.0 -8.910435  70.372582 -594.0 -45.0 -7.0  28.0  1697.0\n",
-       "X4    9200.0 -8.969783  70.030409 -549.0 -45.0 -8.0  27.0  1612.0\n",
-       "X5    9200.0 -9.085326  69.377958 -603.0 -45.0 -8.0  27.0  1437.0\n",
-       "...      ...       ...        ...    ...   ...  ...   ...     ...\n",
-       "X175  9200.0 -9.848587  69.550894 -570.0 -45.0 -9.0  27.0  1958.0\n",
-       "X176  9200.0 -9.620435  70.353607 -594.0 -46.0 -8.0  27.0  2047.0\n",
-       "X177  9200.0 -9.395435  70.934300 -563.0 -45.0 -9.0  27.0  2047.0\n",
-       "X178  9200.0 -9.240435  71.185850 -559.0 -45.0 -8.0  27.0  1915.0\n",
-       "y     9200.0  0.000000   0.000000    0.0   0.0  0.0   0.0     0.0\n",
-       "\n",
-       "[179 rows x 8 columns]"
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# describe non-epileptic data\n",
-    "data[data[\"y\"] == 0].describe().T"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>count</th>\n",
-       "      <th>mean</th>\n",
-       "      <th>std</th>\n",
-       "      <th>min</th>\n",
-       "      <th>25%</th>\n",
-       "      <th>50%</th>\n",
-       "      <th>75%</th>\n",
-       "      <th>max</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>X1</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-21.936522</td>\n",
-       "      <td>342.361939</td>\n",
-       "      <td>-1839.0</td>\n",
-       "      <td>-193.25</td>\n",
-       "      <td>-16.0</td>\n",
-       "      <td>159.00</td>\n",
-       "      <td>1314.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X2</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-19.049130</td>\n",
-       "      <td>343.398782</td>\n",
-       "      <td>-1838.0</td>\n",
-       "      <td>-191.25</td>\n",
-       "      <td>-18.0</td>\n",
-       "      <td>168.25</td>\n",
-       "      <td>1356.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X3</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-15.293913</td>\n",
-       "      <td>337.489643</td>\n",
-       "      <td>-1835.0</td>\n",
-       "      <td>-187.00</td>\n",
-       "      <td>-12.5</td>\n",
-       "      <td>169.25</td>\n",
-       "      <td>1274.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X4</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-9.836087</td>\n",
-       "      <td>332.354833</td>\n",
-       "      <td>-1845.0</td>\n",
-       "      <td>-184.00</td>\n",
-       "      <td>-6.0</td>\n",
-       "      <td>166.25</td>\n",
-       "      <td>1226.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X5</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-3.707391</td>\n",
-       "      <td>332.211163</td>\n",
-       "      <td>-1791.0</td>\n",
-       "      <td>-174.25</td>\n",
-       "      <td>-12.0</td>\n",
-       "      <td>170.00</td>\n",
-       "      <td>1518.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X175</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-25.830870</td>\n",
-       "      <td>339.650467</td>\n",
-       "      <td>-1863.0</td>\n",
-       "      <td>-195.00</td>\n",
-       "      <td>-14.5</td>\n",
-       "      <td>153.25</td>\n",
-       "      <td>1205.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X176</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-25.043913</td>\n",
-       "      <td>335.747017</td>\n",
-       "      <td>-1781.0</td>\n",
-       "      <td>-192.00</td>\n",
-       "      <td>-18.0</td>\n",
-       "      <td>150.00</td>\n",
-       "      <td>1371.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X177</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-24.548261</td>\n",
-       "      <td>335.244512</td>\n",
-       "      <td>-1727.0</td>\n",
-       "      <td>-190.25</td>\n",
-       "      <td>-21.5</td>\n",
-       "      <td>151.25</td>\n",
-       "      <td>1445.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>X178</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>-24.016522</td>\n",
-       "      <td>339.819309</td>\n",
-       "      <td>-1829.0</td>\n",
-       "      <td>-189.00</td>\n",
-       "      <td>-23.0</td>\n",
-       "      <td>157.25</td>\n",
-       "      <td>1380.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>y</th>\n",
-       "      <td>2300.0</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>1.00</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>1.00</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>179 rows × 8 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       count       mean         std     min     25%   50%     75%     max\n",
-       "X1    2300.0 -21.936522  342.361939 -1839.0 -193.25 -16.0  159.00  1314.0\n",
-       "X2    2300.0 -19.049130  343.398782 -1838.0 -191.25 -18.0  168.25  1356.0\n",
-       "X3    2300.0 -15.293913  337.489643 -1835.0 -187.00 -12.5  169.25  1274.0\n",
-       "X4    2300.0  -9.836087  332.354833 -1845.0 -184.00  -6.0  166.25  1226.0\n",
-       "X5    2300.0  -3.707391  332.211163 -1791.0 -174.25 -12.0  170.00  1518.0\n",
-       "...      ...        ...         ...     ...     ...   ...     ...     ...\n",
-       "X175  2300.0 -25.830870  339.650467 -1863.0 -195.00 -14.5  153.25  1205.0\n",
-       "X176  2300.0 -25.043913  335.747017 -1781.0 -192.00 -18.0  150.00  1371.0\n",
-       "X177  2300.0 -24.548261  335.244512 -1727.0 -190.25 -21.5  151.25  1445.0\n",
-       "X178  2300.0 -24.016522  339.819309 -1829.0 -189.00 -23.0  157.25  1380.0\n",
-       "y     2300.0   1.000000    0.000000     1.0    1.00   1.0    1.00     1.0\n",
-       "\n",
-       "[179 rows x 8 columns]"
-      ]
-     },
-     "execution_count": 10,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# describe epileptic data\n",
-    "data[data[\"y\"] == 1].describe().T"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The data in its current form is cumbersome to work with, both for data exploration and for applying machine learning later on. Thus, we reshape our data into a more compliant form.\n",
-    "\n",
-    "First, we reshape our data from its pivoted form into a well-structured table using pandas' [`melt`](https://pandas.pydata.org/docs/reference/api/pandas.melt.html) function. The original index represents each sample, so we preserve it as the `sample_index`. We then extract the number from the `X` column names, which represents a timestamp (we call it `time_index` here) along each window. Finally, we sort the resulting DataFrame so we have a nicely structured table containing each window and corresponding metadata."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# data is in 1s per row format\n",
-    "# first unpivot into single time series, preserve target y, then take the original index, which is the\n",
-    "# \"sample index\" of each sample\n",
-    "data_unpivoted = (\n",
-    "    data.melt(\n",
-    "        id_vars=[\"y\"], var_name=\"time_label\", value_name=\"eeg\", ignore_index=False\n",
-    "    )\n",
-    "    .reset_index()\n",
-    "    .rename(columns={\"index\": \"sample_index\"})\n",
-    ")\n",
-    "\n",
-    "# the time index is the index over the 1s time period in each original row in data\n",
-    "data_unpivoted[\"time_index\"] = (\n",
-    "    data_unpivoted[\"time_label\"].str.extract(r\"(\\d+)\", expand=False).astype(int)\n",
-    ")\n",
-    "\n",
-    "# sort each window according to the sample and time and re-order columns\n",
-    "data_unpivoted = data_unpivoted.sort_values(by=[\"sample_index\", \"time_index\"]).reindex(\n",
-    "    [\"sample_index\", \"time_index\", \"eeg\", \"y\"], axis=1\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Let's have a look at our new DataFrame and the first recording."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>sample_index</th>\n",
-       "      <th>time_index</th>\n",
-       "      <th>eeg</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>135</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>190</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>23000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>229</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>34500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>223</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>192</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2000999</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>174</td>\n",
-       "      <td>5</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2012499</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>175</td>\n",
-       "      <td>4</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2023999</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>176</td>\n",
-       "      <td>-2</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2035499</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>177</td>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2046999</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>178</td>\n",
-       "      <td>20</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>2047000 rows × 4 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "         sample_index  time_index  eeg  y\n",
-       "0                   0           1  135  0\n",
-       "11500               0           2  190  0\n",
-       "23000               0           3  229  0\n",
-       "34500               0           4  223  0\n",
-       "46000               0           5  192  0\n",
-       "...               ...         ...  ... ..\n",
-       "2000999         11499         174    5  0\n",
-       "2012499         11499         175    4  0\n",
-       "2023999         11499         176   -2  0\n",
-       "2035499         11499         177    2  0\n",
-       "2046999         11499         178   20  0\n",
-       "\n",
-       "[2047000 rows x 4 columns]"
-      ]
-     },
-     "execution_count": 12,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_unpivoted"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>sample_index</th>\n",
-       "      <th>time_index</th>\n",
-       "      <th>eeg</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>135</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
-       "      <td>190</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>23000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>3</td>\n",
-       "      <td>229</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>34500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>223</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>192</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1989500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>174</td>\n",
-       "      <td>-103</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2001000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>175</td>\n",
-       "      <td>-127</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2012500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>176</td>\n",
-       "      <td>-116</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024000</th>\n",
-       "      <td>0</td>\n",
-       "      <td>177</td>\n",
-       "      <td>-83</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2035500</th>\n",
-       "      <td>0</td>\n",
-       "      <td>178</td>\n",
-       "      <td>-51</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>178 rows × 4 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "         sample_index  time_index  eeg  y\n",
-       "0                   0           1  135  0\n",
-       "11500               0           2  190  0\n",
-       "23000               0           3  229  0\n",
-       "34500               0           4  223  0\n",
-       "46000               0           5  192  0\n",
-       "...               ...         ...  ... ..\n",
-       "1989500             0         174 -103  0\n",
-       "2001000             0         175 -127  0\n",
-       "2012500             0         176 -116  0\n",
-       "2024000             0         177  -83  0\n",
-       "2035500             0         178  -51  0\n",
-       "\n",
-       "[178 rows x 4 columns]"
-      ]
-     },
-     "execution_count": 13,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_unpivoted[data_unpivoted[\"sample_index\"] == 0]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We then can have a look at some of the EEG signals and get a feel for what we are dealing with. We pick the first `n` (we chose 5, use any number you like) samples of every class and plot the EEG signals side-by-side. "
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "n = 5\n",
-    "\n",
-    "index_n_epileptic = data_unpivoted[data_unpivoted[\"y\"] == 1][\"sample_index\"].unique()[\n",
-    "    :n\n",
-    "]\n",
-    "index_n_nonepileptic = data_unpivoted[data_unpivoted[\"y\"] == 0][\n",
-    "    \"sample_index\"\n",
-    "].unique()[:n]\n",
-    "\n",
-    "samples_to_show = np.concatenate((index_n_epileptic, index_n_nonepileptic))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1106.61x500 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "g = sns.relplot(\n",
-    "    data=data_unpivoted[data_unpivoted[\"sample_index\"].isin(samples_to_show)],\n",
-    "    kind=\"line\",\n",
-    "    x=\"time_index\",\n",
-    "    y=\"eeg\",\n",
-    "    col=\"y\",\n",
-    "    hue=\"sample_index\",\n",
-    "    legend=\"full\",\n",
-    "    palette=sns.color_palette(),\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We can already guess that epileptic signals seem to be a lot more deviating than non-epileptic signals. Let's have a look at the standard deviation of EEG values per class label and compare them:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "<seaborn.axisgrid.FacetGrid at 0x7969f37261a0>"
-      ]
-     },
-     "execution_count": 16,
-     "metadata": {},
-     "output_type": "execute_result"
+    "cells": [
+        {
+            "attachments": {
+                "Linkedin_Optimized_Cover.png": {
+                    "image/png": "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"
+                }
+            },
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "# \n",
+                "![Using getML on EEG data to classify epileptic seizures](attachment:Linkedin_Optimized_Cover.png)"
+            ]
+        },
+        {
+            "attachments": {
+                "Examples2.png": {
+                    "image/png": "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"
+                }
+            },
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Introduction\n",
+                "\n",
+                "### A Primer on Time Series\n",
+                "\n",
+                "Wikipedia defines a time series as follows:\n",
+                "\n",
+                "In mathematics, a time series is a series of data points indexed (or listed or graphed) in time order. Most commonly, a time series is a sequence taken at successive equally spaced points in time.\n",
+                "\n",
+                "The first sentence defines a fundamental feature of time series: your data can be ordered in time. This does not necessitate the existence of a specific time stamp. It only means that data points should be in chronological order.\n",
+                "\n",
+                "Consider the following three examples. All of them are time series. The first example shows force measurements on a robot's arm. We do not know the specific time and date when these measurements were taken. But what we do know is that the data points are indexed according to chronological order. The second example shows hourly traffic volume on interstate 94 (a highway in the United States of America) in the first week of April 2016. This data has an exact date and time when the measurements were taken. The third example is a bio-signal, a recording of an EEG signal of human brain activity. For this kind of data, a specific time stamp can be useful, for example sleep data over longer time periods such as months, but commonly is not necessary for short term bio-signals such as EEG readings or movement data.\n",
+                "\n",
+                "![Examples plots](attachment:Examples2.png)\n",
+                "\n",
+                "The second sentence in Wikipedia's definition says that points in time are most commonly equally spaced. This in fact holds true for most time series data. We know for a fact that traffic volume was measured once every hour. Although we do not have a time stamp for the robot arm data, we can assume that the data recording sensor measured in equally spaced time intervals. \n",
+                "\n",
+                "As a matter of fact, even these equidistant time intervals might be an assumption in data sets like the above examples, as all sensors physically record data points with some form of timely variance due to multiple reasons such as sensor or data saving lags and so on. Additionally, there might be an inherent variance; for example a person never sleeps at exactly the same time every day. However, in many real-world data sets you will find that the assumption of equally spaced data points in time is violated and you will have to deal with that."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Time Series Prediction Tasks\n",
+                "\n",
+                "There are several kinds of time series problems, but they can be roughly subdivided into two kinds: *Forecasts* and *Nowcasts*.\n",
+                "\n",
+                "- A *forecast* tries to predict the future. For instance, when dealing with the Interstate 94 data, you will most likely want to predict the traffic volume in one hour from now or one week from now. This time period is often called the *forecast horizon*.\n",
+                "\n",
+                "- A *nowcast* tries to predict another time series or class-label based on some time series. For instance, when confronted with the robot data, you might want to predict the force vector based on other sensor data. This is useful when other sensor data are easier to measure than the force vector. On the other hand, a very common task for bio-signals is classifying some state based on these signals, such as sleep quality or epileptic seizure occurance.\n",
+                "\n",
+                "A very interesting question is also whether you are allowed to use lagged targets. For instance, if you want to predict the traffic volume for the next hour, are you allowed to use the traffic volume for the last three hours? In this case, probably yes, but if you want to do a nowcast on the force vector of the robot's arm, probably no, because the entire idea is that measuring the force vector is difficult and therefore we cannot assume that we have the force vector data of the recent past when we are making a prediction. This question is also very relevant for classification tasks, as you face the question of how far in the past data is relevant to classifying a specific time point or interval. For example, in sleep data classification, events that occurred several days, weeks, or even months in the past might have an influence, while an epileptic seizure might be detectable in the short term."
+            ]
+        },
+        {
+            "attachments": {
+                "ManualFeatureEngineering2.png": {
+                    "image/png": "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"
+                }
+            },
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Why We Need Feature Engineering\n",
+                "\n",
+                "The next question is: why do we need feature engineering in the first place? Isn't machine learning supposed to magically extract the information it needs from the data? Unfortunately, it is not quite that simple. Standard machine learning algorithms require the data to be presented in a *flat table*. \n",
+                "\n",
+                "We say *standard machine learning* and that means that there are exceptions. *Deep learning* can automatically extract features from unstructured data like images, text and speech. On the other hand, *relational learning* can automatically extract features from structured relational data.\n",
+                "\n",
+                "In the absence of these two approaches, we have to manually engineer features. This involves interactions between *data scientists*, who understand statistics, and *domain experts*, who understand the underlying problem domain. For example, domain experts provide domain know-how to data scientists, who write code and software that needs to be evaluated and optimized by both parties. The result is a flat table containing features to be used in machine learning. However, the process of getting to that table is cumbersome, time consuming and error prone.\n",
+                "\n",
+                "![Manual Feature Engineering](attachment:ManualFeatureEngineering2.png)\n",
+                "\n",
+                "In this figure, we refer to *relational data*. Relational data is structured data that you would usually find in relational databases such as PostgreSQL, MySQL, MariaDB, BigQuery or Redshift. However, the same holds true for time series. In fact, time series can be considered a special case of relational data. Below, we will see why that is."
+            ]
+        },
+        {
+            "attachments": {
+                "FeatureEngineeringExample2.png": {
+                    "image/png": "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"
+                }
+            },
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Interesting Features for Time Series\n",
+                "\n",
+                "Two important characteristics are *trend* and *seasonality*.\n",
+                "\n",
+                "- The *trend* indicates whether your time series increases or decreases over time.\n",
+                "\n",
+                "- The *seasonality* indicates any patterns that vary by hour, day etc. For instance, the Interstate 94 dataset has very obvious seasonal patterns: Traffic volume is highest in the morning (when people drive to work) and in the evening (when people get home from work), very low at night and overall lower on weekends.\n",
+                "\n",
+                "But there are many other important characteristics, such as the *average*, *median*, *max*, *min* and *standard deviation* of your observations over time. So when you build features for time series, a very common approach is to use a *sliding window* technique. Whenever you want to make a prediction you take all the values over a fixed period of time right up to when you want to make the prediction and then apply all sorts of aggregations to this. For classification tasks, every window or time point is classified by a certain label.\n",
+                "\n",
+                "![Feature Engineering Example 2](attachment:FeatureEngineeringExample2.png)\n",
+                "\n",
+                "This depiction shows how we can engineer a very simple feature for a time series. As we would for relational data, we define some kind of criterion over which we identify the data we are interested in (in this case, the last five days), which we then aggregate using some aggregation function (in this case, the average). This is a very simple example how feature engineering for time series usually works. But if you engineer features for time series in this way, you are effectively thinking of time series as relational data: You are identifying relevant data from your data set and aggregating it, just like you would for relational data. In fact, what we are doing is effectively a *self join*, because we are joining a table to itself.\n",
+                "\n",
+                "This is just a simple example. But there are many more features you can generate. For instance, for the EEG data, you build features like this:\n",
+                "\n",
+                "- Average EEG value in the last second\n",
+                "- Maximum EEG value in the last second\n",
+                "- Minimum EEG value in the last second\n",
+                "- Median EEG value in the last second\n",
+                "- First measurement of EEG value in the last second\n",
+                "- Last measurement of EEG value in the last second\n",
+                "- Variance of the EEG value in the last second\n",
+                "- Standard deviation of the EEG value in the last second\n",
+                "- Exponentially weighted moving average of the EEG value in the last second\n",
+                "- 1%-quantile of the EEG value in the last second\n",
+                "- 5%-quantile of the EEG value in the last second\n",
+                "- ...\n",
+                "\n",
+                "Of course, we mustn't just assume that one second is the right period of time to use. So we could take all of these features and calculate them for the last minute or last three hours.\n",
+                "\n",
+                "Moreover, the features we have discussed so far don't really take seasonality into account (with the exception of first measurement of EEG value in the last second, because this is the measurement from exactly one second ago).\n",
+                "\n",
+                "On the other hand, the Interstate 94 dataset is strongly seasonal, we should take that into account as well. So we could calculate features like this:\n",
+                "\n",
+                "- Average traffic volume in the last four weeks, but only where weekday equals the weekday the point in time we want to predict.\n",
+                "- Average traffic volume in the last four weeks, but only where hour of the day equals the hour of the point in time we want to predict.\n",
+                "- Average traffic volume in the last four weeks, but only where both the weekday and the hour of the day equal the point in time we want to predict.\n",
+                "- Maximum traffic volume in the last four weeks, but only where weekday equals the weekday the point in time we want to predict.\n",
+                "- ...\n",
+                "\n",
+                "What should be very obvious at this point is that there are *many* features that you can generate like this, even for a very simple time series problem. When you have a multivariate time series (meaning you have more than one input variable), you can apply these techniques to every single column in your input data and you will get many, many features. You would very likely have to apply some kind of feature selection techniques to focus on the most useful features.\n",
+                "\n",
+                "The beauty of this approach is that it is very flexible and uses few assumptions. We have noted above that many time series are *not* equally-spaced. This would be a problem for classical time series analyses like ARIMA or ARMA. Not here. Nothing about this approach makes any assumptions about the spacing. "
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Automated Feature Engineering for Time Series\n",
+                "\n",
+                "You will find a lot of examples where people conduct this work manually (mainly using pandas). But today, we are not limited to manual feature engineering. There are numerous tools and libraries which can automate away this kind of work.\n",
+                "\n",
+                "Unfortunately, automated feature engineering has been getting a bit of a bad rap, mainly for taking too long. And it isn't wrong: Some of the more well-known libraries like featuretools or tsfresh are slow and not very memory-efficient.\n",
+                "\n",
+                "Overall, the features extracted from time series by such libraries are quite similar [[Henderson & Fulcher](https://ieeexplore.ieee.org/document/9679937)]. However, the stark differences in terms of runtime and memory consumption make it worthwhile taking a closer look as some of the newer tools and libraries like getML or tsflex are highly optimized and can generate many features in a short period of time. \n",
+                "\n",
+                "In the following, we would like to introduce time series classification using the getML Pyhon API. Above, we have discussed how time series can be seen as a form of relational data and introduced the term relational learning. We can utilize a very simple relational learning approach by interpreting the time series as a form of relational data and conduct a self join."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### An Introduction to Univariate Time Series Classification with getML\n",
+                "\n",
+                "In this tutorial, you will learn how to use getML to classify univariate time series. We first explore the data and learn what we are dealing with. Subsequently, we demonstrate how to efficiently build a full fledged machine learning data model and how to use getML's automatic feature learning algorithm FastProp.\n",
+                "\n",
+                "### About the Dataset\n",
+                "The original dataset from the reference comprises 500 files, each file representing a single person/subject. Each recording contains the EEG signal value of brain activity for 23.6s sampled into 4096 data points. These recordings have been split into 1s windows. This results in 23 x 500 = 11500 windows of EEG data over time in 178 datapoints and each window is categorized into 5 labels:\n",
+                "\n",
+                "1. Seizure activity\n",
+                "2. EEG recorded at tumor site\n",
+                "3. EEG recorded in healthy brain area\n",
+                "4. eyes closed during recording\n",
+                "5. eyes open during recording\n",
+                "\n",
+                "Subjects labeled with classes 2-5 did not have epileptic seizures. We can thus do a binary classification of subjects suffering an epileptic seizure or not, meaning classes 1 or 0, respectively.\n",
+                "\n",
+                "### Acknowledgements\n",
+                "Andrzejak RG, Lehnertz K, Rieke C, Mormann F, David P, Elger CE (2001) Indications of nonlinear deterministic and finite dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state, Phys. Rev. E, 64, 061907"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Setup\n",
+                "\n",
+                "### Start up getML\n",
+                "\n",
+                "First, we import the necessary libraries  and launch the [getML engine](https://docs.getml.com/latest/user_guide/getml_suite/engine.html). The engine runs in the background and takes care of all the heavy lifting for you. This includes things like our powerful database engine and efficient algorithms as well as the [getML monitor](https://docs.getml.com/latest/user_guide/getml_suite/monitor.html), which you can access by pointing your browser to http://localhost:1709/#/"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 1,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "\n",
+                        "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
+                        "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
+                        "Note: you may need to restart the kernel to use updated packages.\n"
+                    ]
+                }
+            ],
+            "source": [
+                "%pip install -q \"getml==1.4.0\" \"numpy<2.0.0\" \"matplotlib~=3.9\" \"seaborn~=0.13\""
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 2,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "import numpy as np\n",
+                "import pandas as pd\n",
+                "\n",
+                "import matplotlib.pyplot as plt\n",
+                "import seaborn as sns\n",
+                "\n",
+                "sns.set_style(\"whitegrid\")\n",
+                "\n",
+                "import getml"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "All your work is organized into [projects](https://docs.getml.com/latest/user_guide/project_management/project_management.html). You can easily set any name for your current project. The engine will create a new project or use an existing one if the project name already exists. It will also provide you with a direct link to the project within the monitor."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 3,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Launching ./getML --allow-push-notifications=true --allow-remote-ips=false --home-directory=/home/user/.local/share/hatch/env/virtual/getml-demo/txflr3_Z/getml-demo/lib/python3.10/site-packages/getml --in-memory=true --install=false --launch-browser=true --log=false in /home/user/.local/share/hatch/env/virtual/getml-demo/txflr3_Z/getml-demo/lib/python3.10/site-packages/getml/.getML/getml-1.4.0-x64-community-edition-linux...\n",
+                        "Launched the getML engine. The log output will be stored in /home/user/.getML/logs/20240826151418.log.\n",
+                        "Loading pipelines... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "\n",
+                        "Connected to project 'epilepsy_recognition'\n"
+                    ]
+                }
+            ],
+            "source": [
+                "getml.engine.launch()\n",
+                "getml.engine.set_project(\"epilepsy_recognition\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "You can manage your projects conveniently through the monitor web interface or directly using Python commands. For example, you can suspend your current project to free resources using [`getml.project.suspend()`](https://docs.getml.com/latest/api/project/getml.project.suspend.html), switch to another project on the fly using [`getml.project.switch('new project name')`](https://docs.getml.com/latest/api/project/getml.project.switch.html), or restart using [`getml.project.restart()`](https://docs.getml.com/latest/api/project/getml.project.restart.html) should something go wrong. You can even save your current project to disk using [`getml.project.save('filename')`](https://docs.getml.com/latest/api/project/getml.project.save.html) and load it with [`getml.project.load('filename')`](https://docs.getml.com/latest/api/project/getml.project.load.html)."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Load Data\n",
+                "\n",
+                "The original dataset was hosted on the [UCI repository](https://archive.ics.uci.edu/ml/datasets/Epileptic+Seizure+Recognition) but was unfortunately removed. You can get the dataset via Kaggle, [here](https://www.kaggle.com/datasets/harunshimanto/epileptic-seizure-recognition).\n",
+                "\n",
+                "The dataset we will be working on is stored in a CSV file located on disk. As we will perform data exploration, we will first load the data into a pandas DataFrame, as usual, and examine the raw data."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 4,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>Unnamed</th>\n",
+                            "      <th>X1</th>\n",
+                            "      <th>X2</th>\n",
+                            "      <th>X3</th>\n",
+                            "      <th>X4</th>\n",
+                            "      <th>X5</th>\n",
+                            "      <th>X6</th>\n",
+                            "      <th>X7</th>\n",
+                            "      <th>X8</th>\n",
+                            "      <th>X9</th>\n",
+                            "      <th>...</th>\n",
+                            "      <th>X170</th>\n",
+                            "      <th>X171</th>\n",
+                            "      <th>X172</th>\n",
+                            "      <th>X173</th>\n",
+                            "      <th>X174</th>\n",
+                            "      <th>X175</th>\n",
+                            "      <th>X176</th>\n",
+                            "      <th>X177</th>\n",
+                            "      <th>X178</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>X21.V1.791</td>\n",
+                            "      <td>135</td>\n",
+                            "      <td>190</td>\n",
+                            "      <td>229</td>\n",
+                            "      <td>223</td>\n",
+                            "      <td>192</td>\n",
+                            "      <td>125</td>\n",
+                            "      <td>55</td>\n",
+                            "      <td>-9</td>\n",
+                            "      <td>-33</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>-17</td>\n",
+                            "      <td>-15</td>\n",
+                            "      <td>-31</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-103</td>\n",
+                            "      <td>-127</td>\n",
+                            "      <td>-116</td>\n",
+                            "      <td>-83</td>\n",
+                            "      <td>-51</td>\n",
+                            "      <td>4</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>1</th>\n",
+                            "      <td>X15.V1.924</td>\n",
+                            "      <td>386</td>\n",
+                            "      <td>382</td>\n",
+                            "      <td>356</td>\n",
+                            "      <td>331</td>\n",
+                            "      <td>320</td>\n",
+                            "      <td>315</td>\n",
+                            "      <td>307</td>\n",
+                            "      <td>272</td>\n",
+                            "      <td>244</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>164</td>\n",
+                            "      <td>150</td>\n",
+                            "      <td>146</td>\n",
+                            "      <td>152</td>\n",
+                            "      <td>157</td>\n",
+                            "      <td>156</td>\n",
+                            "      <td>154</td>\n",
+                            "      <td>143</td>\n",
+                            "      <td>129</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2</th>\n",
+                            "      <td>X8.V1.1</td>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-39</td>\n",
+                            "      <td>-47</td>\n",
+                            "      <td>-37</td>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-36</td>\n",
+                            "      <td>-57</td>\n",
+                            "      <td>-73</td>\n",
+                            "      <td>-85</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>57</td>\n",
+                            "      <td>64</td>\n",
+                            "      <td>48</td>\n",
+                            "      <td>19</td>\n",
+                            "      <td>-12</td>\n",
+                            "      <td>-30</td>\n",
+                            "      <td>-35</td>\n",
+                            "      <td>-35</td>\n",
+                            "      <td>-36</td>\n",
+                            "      <td>5</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>3</th>\n",
+                            "      <td>X16.V1.60</td>\n",
+                            "      <td>-105</td>\n",
+                            "      <td>-101</td>\n",
+                            "      <td>-96</td>\n",
+                            "      <td>-92</td>\n",
+                            "      <td>-89</td>\n",
+                            "      <td>-95</td>\n",
+                            "      <td>-102</td>\n",
+                            "      <td>-100</td>\n",
+                            "      <td>-87</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>-82</td>\n",
+                            "      <td>-81</td>\n",
+                            "      <td>-80</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-85</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-72</td>\n",
+                            "      <td>-69</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>5</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>4</th>\n",
+                            "      <td>X20.V1.54</td>\n",
+                            "      <td>-9</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>-98</td>\n",
+                            "      <td>-102</td>\n",
+                            "      <td>-78</td>\n",
+                            "      <td>-48</td>\n",
+                            "      <td>-16</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>4</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>-12</td>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-41</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>-83</td>\n",
+                            "      <td>-89</td>\n",
+                            "      <td>-73</td>\n",
+                            "      <td>5</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>5 rows × 180 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "      Unnamed   X1   X2   X3   X4   X5   X6   X7   X8   X9  ...  X170  X171  \\\n",
+                            "0  X21.V1.791  135  190  229  223  192  125   55   -9  -33  ...   -17   -15   \n",
+                            "1  X15.V1.924  386  382  356  331  320  315  307  272  244  ...   164   150   \n",
+                            "2     X8.V1.1  -32  -39  -47  -37  -32  -36  -57  -73  -85  ...    57    64   \n",
+                            "3   X16.V1.60 -105 -101  -96  -92  -89  -95 -102 -100  -87  ...   -82   -81   \n",
+                            "4   X20.V1.54   -9  -65  -98 -102  -78  -48  -16    0  -21  ...     4     2   \n",
+                            "\n",
+                            "   X172  X173  X174  X175  X176  X177  X178  y  \n",
+                            "0   -31   -77  -103  -127  -116   -83   -51  4  \n",
+                            "1   146   152   157   156   154   143   129  1  \n",
+                            "2    48    19   -12   -30   -35   -35   -36  5  \n",
+                            "3   -80   -77   -85   -77   -72   -69   -65  5  \n",
+                            "4   -12   -32   -41   -65   -83   -89   -73  5  \n",
+                            "\n",
+                            "[5 rows x 180 columns]"
+                        ]
+                    },
+                    "execution_count": 4,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data = pd.read_csv(\"data/Epileptic Seizure Recognition.csv\")\n",
+                "\n",
+                "# view first 5 rows of the data\n",
+                "data.head()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The first column contains extraneous metadata and can be dropped. The last column is named `y` and contains the class labels. As described above, we will do a binary classification into epileptic seizure (label 1) or not (label 2-5). Thus, we can set the labels 2-5 to 0, representing a non-epileptic instance, and 1 for epileptic seizure."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 5,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "data.drop(\"Unnamed\", axis=1, inplace=True)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 6,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# classify having epileptic seizure or not\n",
+                "class_relabeling = {1: 1, 2: 0, 3: 0, 4: 0, 5: 0}\n",
+                "data.replace({\"y\": class_relabeling}, inplace=True)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now we can check which values we have in the label column and the DataFrame in general."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 7,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Number of records epileptic 2300 vs non-epileptic 9200\n"
+                    ]
+                }
+            ],
+            "source": [
+                "counts = data[\"y\"].value_counts()\n",
+                "print(f\"Number of records epileptic {counts[1]} vs non-epileptic {counts[0]}\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 8,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>X1</th>\n",
+                            "      <th>X2</th>\n",
+                            "      <th>X3</th>\n",
+                            "      <th>X4</th>\n",
+                            "      <th>X5</th>\n",
+                            "      <th>X6</th>\n",
+                            "      <th>X7</th>\n",
+                            "      <th>X8</th>\n",
+                            "      <th>X9</th>\n",
+                            "      <th>X10</th>\n",
+                            "      <th>...</th>\n",
+                            "      <th>X170</th>\n",
+                            "      <th>X171</th>\n",
+                            "      <th>X172</th>\n",
+                            "      <th>X173</th>\n",
+                            "      <th>X174</th>\n",
+                            "      <th>X175</th>\n",
+                            "      <th>X176</th>\n",
+                            "      <th>X177</th>\n",
+                            "      <th>X178</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>135</td>\n",
+                            "      <td>190</td>\n",
+                            "      <td>229</td>\n",
+                            "      <td>223</td>\n",
+                            "      <td>192</td>\n",
+                            "      <td>125</td>\n",
+                            "      <td>55</td>\n",
+                            "      <td>-9</td>\n",
+                            "      <td>-33</td>\n",
+                            "      <td>-38</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>-17</td>\n",
+                            "      <td>-15</td>\n",
+                            "      <td>-31</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-103</td>\n",
+                            "      <td>-127</td>\n",
+                            "      <td>-116</td>\n",
+                            "      <td>-83</td>\n",
+                            "      <td>-51</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>1</th>\n",
+                            "      <td>386</td>\n",
+                            "      <td>382</td>\n",
+                            "      <td>356</td>\n",
+                            "      <td>331</td>\n",
+                            "      <td>320</td>\n",
+                            "      <td>315</td>\n",
+                            "      <td>307</td>\n",
+                            "      <td>272</td>\n",
+                            "      <td>244</td>\n",
+                            "      <td>232</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>164</td>\n",
+                            "      <td>150</td>\n",
+                            "      <td>146</td>\n",
+                            "      <td>152</td>\n",
+                            "      <td>157</td>\n",
+                            "      <td>156</td>\n",
+                            "      <td>154</td>\n",
+                            "      <td>143</td>\n",
+                            "      <td>129</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2</th>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-39</td>\n",
+                            "      <td>-47</td>\n",
+                            "      <td>-37</td>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-36</td>\n",
+                            "      <td>-57</td>\n",
+                            "      <td>-73</td>\n",
+                            "      <td>-85</td>\n",
+                            "      <td>-94</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>57</td>\n",
+                            "      <td>64</td>\n",
+                            "      <td>48</td>\n",
+                            "      <td>19</td>\n",
+                            "      <td>-12</td>\n",
+                            "      <td>-30</td>\n",
+                            "      <td>-35</td>\n",
+                            "      <td>-35</td>\n",
+                            "      <td>-36</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>3</th>\n",
+                            "      <td>-105</td>\n",
+                            "      <td>-101</td>\n",
+                            "      <td>-96</td>\n",
+                            "      <td>-92</td>\n",
+                            "      <td>-89</td>\n",
+                            "      <td>-95</td>\n",
+                            "      <td>-102</td>\n",
+                            "      <td>-100</td>\n",
+                            "      <td>-87</td>\n",
+                            "      <td>-79</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>-82</td>\n",
+                            "      <td>-81</td>\n",
+                            "      <td>-80</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-85</td>\n",
+                            "      <td>-77</td>\n",
+                            "      <td>-72</td>\n",
+                            "      <td>-69</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>4</th>\n",
+                            "      <td>-9</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>-98</td>\n",
+                            "      <td>-102</td>\n",
+                            "      <td>-78</td>\n",
+                            "      <td>-48</td>\n",
+                            "      <td>-16</td>\n",
+                            "      <td>0</td>\n",
+                            "      <td>-21</td>\n",
+                            "      <td>-59</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>4</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>-12</td>\n",
+                            "      <td>-32</td>\n",
+                            "      <td>-41</td>\n",
+                            "      <td>-65</td>\n",
+                            "      <td>-83</td>\n",
+                            "      <td>-89</td>\n",
+                            "      <td>-73</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>5 rows × 179 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "    X1   X2   X3   X4   X5   X6   X7   X8   X9  X10  ...  X170  X171  X172  \\\n",
+                            "0  135  190  229  223  192  125   55   -9  -33  -38  ...   -17   -15   -31   \n",
+                            "1  386  382  356  331  320  315  307  272  244  232  ...   164   150   146   \n",
+                            "2  -32  -39  -47  -37  -32  -36  -57  -73  -85  -94  ...    57    64    48   \n",
+                            "3 -105 -101  -96  -92  -89  -95 -102 -100  -87  -79  ...   -82   -81   -80   \n",
+                            "4   -9  -65  -98 -102  -78  -48  -16    0  -21  -59  ...     4     2   -12   \n",
+                            "\n",
+                            "   X173  X174  X175  X176  X177  X178  y  \n",
+                            "0   -77  -103  -127  -116   -83   -51  0  \n",
+                            "1   152   157   156   154   143   129  1  \n",
+                            "2    19   -12   -30   -35   -35   -36  0  \n",
+                            "3   -77   -85   -77   -72   -69   -65  0  \n",
+                            "4   -32   -41   -65   -83   -89   -73  0  \n",
+                            "\n",
+                            "[5 rows x 179 columns]"
+                        ]
+                    },
+                    "execution_count": 8,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data.head()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### Explore Data\n",
+                "\n",
+                "We first have a look at some common statistics of our data divided into both classes."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 9,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>count</th>\n",
+                            "      <th>mean</th>\n",
+                            "      <th>std</th>\n",
+                            "      <th>min</th>\n",
+                            "      <th>25%</th>\n",
+                            "      <th>50%</th>\n",
+                            "      <th>75%</th>\n",
+                            "      <th>max</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>X1</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-8.992609</td>\n",
+                            "      <td>70.455286</td>\n",
+                            "      <td>-566.0</td>\n",
+                            "      <td>-44.0</td>\n",
+                            "      <td>-7.0</td>\n",
+                            "      <td>26.0</td>\n",
+                            "      <td>1726.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X2</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-8.877174</td>\n",
+                            "      <td>70.560110</td>\n",
+                            "      <td>-609.0</td>\n",
+                            "      <td>-44.0</td>\n",
+                            "      <td>-7.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>1713.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X3</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-8.910435</td>\n",
+                            "      <td>70.372582</td>\n",
+                            "      <td>-594.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-7.0</td>\n",
+                            "      <td>28.0</td>\n",
+                            "      <td>1697.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X4</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-8.969783</td>\n",
+                            "      <td>70.030409</td>\n",
+                            "      <td>-549.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-8.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>1612.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X5</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-9.085326</td>\n",
+                            "      <td>69.377958</td>\n",
+                            "      <td>-603.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-8.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>1437.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X175</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-9.848587</td>\n",
+                            "      <td>69.550894</td>\n",
+                            "      <td>-570.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-9.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>1958.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X176</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-9.620435</td>\n",
+                            "      <td>70.353607</td>\n",
+                            "      <td>-594.0</td>\n",
+                            "      <td>-46.0</td>\n",
+                            "      <td>-8.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>2047.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X177</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-9.395435</td>\n",
+                            "      <td>70.934300</td>\n",
+                            "      <td>-563.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-9.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>2047.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X178</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>-9.240435</td>\n",
+                            "      <td>71.185850</td>\n",
+                            "      <td>-559.0</td>\n",
+                            "      <td>-45.0</td>\n",
+                            "      <td>-8.0</td>\n",
+                            "      <td>27.0</td>\n",
+                            "      <td>1915.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>y</th>\n",
+                            "      <td>9200.0</td>\n",
+                            "      <td>0.000000</td>\n",
+                            "      <td>0.000000</td>\n",
+                            "      <td>0.0</td>\n",
+                            "      <td>0.0</td>\n",
+                            "      <td>0.0</td>\n",
+                            "      <td>0.0</td>\n",
+                            "      <td>0.0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>179 rows × 8 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "       count      mean        std    min   25%  50%   75%     max\n",
+                            "X1    9200.0 -8.992609  70.455286 -566.0 -44.0 -7.0  26.0  1726.0\n",
+                            "X2    9200.0 -8.877174  70.560110 -609.0 -44.0 -7.0  27.0  1713.0\n",
+                            "X3    9200.0 -8.910435  70.372582 -594.0 -45.0 -7.0  28.0  1697.0\n",
+                            "X4    9200.0 -8.969783  70.030409 -549.0 -45.0 -8.0  27.0  1612.0\n",
+                            "X5    9200.0 -9.085326  69.377958 -603.0 -45.0 -8.0  27.0  1437.0\n",
+                            "...      ...       ...        ...    ...   ...  ...   ...     ...\n",
+                            "X175  9200.0 -9.848587  69.550894 -570.0 -45.0 -9.0  27.0  1958.0\n",
+                            "X176  9200.0 -9.620435  70.353607 -594.0 -46.0 -8.0  27.0  2047.0\n",
+                            "X177  9200.0 -9.395435  70.934300 -563.0 -45.0 -9.0  27.0  2047.0\n",
+                            "X178  9200.0 -9.240435  71.185850 -559.0 -45.0 -8.0  27.0  1915.0\n",
+                            "y     9200.0  0.000000   0.000000    0.0   0.0  0.0   0.0     0.0\n",
+                            "\n",
+                            "[179 rows x 8 columns]"
+                        ]
+                    },
+                    "execution_count": 9,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "# describe non-epileptic data\n",
+                "data[data[\"y\"] == 0].describe().T"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 10,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>count</th>\n",
+                            "      <th>mean</th>\n",
+                            "      <th>std</th>\n",
+                            "      <th>min</th>\n",
+                            "      <th>25%</th>\n",
+                            "      <th>50%</th>\n",
+                            "      <th>75%</th>\n",
+                            "      <th>max</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>X1</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-21.936522</td>\n",
+                            "      <td>342.361939</td>\n",
+                            "      <td>-1839.0</td>\n",
+                            "      <td>-193.25</td>\n",
+                            "      <td>-16.0</td>\n",
+                            "      <td>159.00</td>\n",
+                            "      <td>1314.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X2</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-19.049130</td>\n",
+                            "      <td>343.398782</td>\n",
+                            "      <td>-1838.0</td>\n",
+                            "      <td>-191.25</td>\n",
+                            "      <td>-18.0</td>\n",
+                            "      <td>168.25</td>\n",
+                            "      <td>1356.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X3</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-15.293913</td>\n",
+                            "      <td>337.489643</td>\n",
+                            "      <td>-1835.0</td>\n",
+                            "      <td>-187.00</td>\n",
+                            "      <td>-12.5</td>\n",
+                            "      <td>169.25</td>\n",
+                            "      <td>1274.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X4</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-9.836087</td>\n",
+                            "      <td>332.354833</td>\n",
+                            "      <td>-1845.0</td>\n",
+                            "      <td>-184.00</td>\n",
+                            "      <td>-6.0</td>\n",
+                            "      <td>166.25</td>\n",
+                            "      <td>1226.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X5</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-3.707391</td>\n",
+                            "      <td>332.211163</td>\n",
+                            "      <td>-1791.0</td>\n",
+                            "      <td>-174.25</td>\n",
+                            "      <td>-12.0</td>\n",
+                            "      <td>170.00</td>\n",
+                            "      <td>1518.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X175</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-25.830870</td>\n",
+                            "      <td>339.650467</td>\n",
+                            "      <td>-1863.0</td>\n",
+                            "      <td>-195.00</td>\n",
+                            "      <td>-14.5</td>\n",
+                            "      <td>153.25</td>\n",
+                            "      <td>1205.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X176</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-25.043913</td>\n",
+                            "      <td>335.747017</td>\n",
+                            "      <td>-1781.0</td>\n",
+                            "      <td>-192.00</td>\n",
+                            "      <td>-18.0</td>\n",
+                            "      <td>150.00</td>\n",
+                            "      <td>1371.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X177</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-24.548261</td>\n",
+                            "      <td>335.244512</td>\n",
+                            "      <td>-1727.0</td>\n",
+                            "      <td>-190.25</td>\n",
+                            "      <td>-21.5</td>\n",
+                            "      <td>151.25</td>\n",
+                            "      <td>1445.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>X178</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>-24.016522</td>\n",
+                            "      <td>339.819309</td>\n",
+                            "      <td>-1829.0</td>\n",
+                            "      <td>-189.00</td>\n",
+                            "      <td>-23.0</td>\n",
+                            "      <td>157.25</td>\n",
+                            "      <td>1380.0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>y</th>\n",
+                            "      <td>2300.0</td>\n",
+                            "      <td>1.000000</td>\n",
+                            "      <td>0.000000</td>\n",
+                            "      <td>1.0</td>\n",
+                            "      <td>1.00</td>\n",
+                            "      <td>1.0</td>\n",
+                            "      <td>1.00</td>\n",
+                            "      <td>1.0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>179 rows × 8 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "       count       mean         std     min     25%   50%     75%     max\n",
+                            "X1    2300.0 -21.936522  342.361939 -1839.0 -193.25 -16.0  159.00  1314.0\n",
+                            "X2    2300.0 -19.049130  343.398782 -1838.0 -191.25 -18.0  168.25  1356.0\n",
+                            "X3    2300.0 -15.293913  337.489643 -1835.0 -187.00 -12.5  169.25  1274.0\n",
+                            "X4    2300.0  -9.836087  332.354833 -1845.0 -184.00  -6.0  166.25  1226.0\n",
+                            "X5    2300.0  -3.707391  332.211163 -1791.0 -174.25 -12.0  170.00  1518.0\n",
+                            "...      ...        ...         ...     ...     ...   ...     ...     ...\n",
+                            "X175  2300.0 -25.830870  339.650467 -1863.0 -195.00 -14.5  153.25  1205.0\n",
+                            "X176  2300.0 -25.043913  335.747017 -1781.0 -192.00 -18.0  150.00  1371.0\n",
+                            "X177  2300.0 -24.548261  335.244512 -1727.0 -190.25 -21.5  151.25  1445.0\n",
+                            "X178  2300.0 -24.016522  339.819309 -1829.0 -189.00 -23.0  157.25  1380.0\n",
+                            "y     2300.0   1.000000    0.000000     1.0    1.00   1.0    1.00     1.0\n",
+                            "\n",
+                            "[179 rows x 8 columns]"
+                        ]
+                    },
+                    "execution_count": 10,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "# describe epileptic data\n",
+                "data[data[\"y\"] == 1].describe().T"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The data in its current form is cumbersome to work with, both for data exploration and for applying machine learning later on. Thus, we reshape our data into a more compliant form.\n",
+                "\n",
+                "First, we reshape our data from its pivoted form into a well-structured table using pandas' [`melt`](https://pandas.pydata.org/docs/reference/api/pandas.melt.html) function. The original index represents each sample, so we preserve it as the `sample_index`. We then extract the number from the `X` column names, which represents a timestamp (we call it `time_index` here) along each window. Finally, we sort the resulting DataFrame so we have a nicely structured table containing each window and corresponding metadata."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 11,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "# data is in 1s per row format\n",
+                "# first unpivot into single time series, preserve target y, then take the original index, which is the\n",
+                "# \"sample index\" of each sample\n",
+                "data_unpivoted = (\n",
+                "    data.melt(\n",
+                "        id_vars=[\"y\"], var_name=\"time_label\", value_name=\"eeg\", ignore_index=False\n",
+                "    )\n",
+                "    .reset_index()\n",
+                "    .rename(columns={\"index\": \"sample_index\"})\n",
+                ")\n",
+                "\n",
+                "# the time index is the index over the 1s time period in each original row in data\n",
+                "data_unpivoted[\"time_index\"] = (\n",
+                "    data_unpivoted[\"time_label\"].str.extract(r\"(\\d+)\", expand=False).astype(int)\n",
+                ")\n",
+                "\n",
+                "# sort each window according to the sample and time and re-order columns\n",
+                "data_unpivoted = data_unpivoted.sort_values(by=[\"sample_index\", \"time_index\"]).reindex(\n",
+                "    [\"sample_index\", \"time_index\", \"eeg\", \"y\"], axis=1\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Let's have a look at our new DataFrame and the first recording."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 12,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>sample_index</th>\n",
+                            "      <th>time_index</th>\n",
+                            "      <th>eeg</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>1</td>\n",
+                            "      <td>135</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>190</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>23000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>3</td>\n",
+                            "      <td>229</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>34500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>4</td>\n",
+                            "      <td>223</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>46000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>5</td>\n",
+                            "      <td>192</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2000999</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>174</td>\n",
+                            "      <td>5</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2012499</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>175</td>\n",
+                            "      <td>4</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2023999</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>176</td>\n",
+                            "      <td>-2</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2035499</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>177</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2046999</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>178</td>\n",
+                            "      <td>20</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>2047000 rows × 4 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "         sample_index  time_index  eeg  y\n",
+                            "0                   0           1  135  0\n",
+                            "11500               0           2  190  0\n",
+                            "23000               0           3  229  0\n",
+                            "34500               0           4  223  0\n",
+                            "46000               0           5  192  0\n",
+                            "...               ...         ...  ... ..\n",
+                            "2000999         11499         174    5  0\n",
+                            "2012499         11499         175    4  0\n",
+                            "2023999         11499         176   -2  0\n",
+                            "2035499         11499         177    2  0\n",
+                            "2046999         11499         178   20  0\n",
+                            "\n",
+                            "[2047000 rows x 4 columns]"
+                        ]
+                    },
+                    "execution_count": 12,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_unpivoted"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 13,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>sample_index</th>\n",
+                            "      <th>time_index</th>\n",
+                            "      <th>eeg</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>1</td>\n",
+                            "      <td>135</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>2</td>\n",
+                            "      <td>190</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>23000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>3</td>\n",
+                            "      <td>229</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>34500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>4</td>\n",
+                            "      <td>223</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>46000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>5</td>\n",
+                            "      <td>192</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>1989500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>174</td>\n",
+                            "      <td>-103</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2001000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>175</td>\n",
+                            "      <td>-127</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2012500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>176</td>\n",
+                            "      <td>-116</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2024000</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>177</td>\n",
+                            "      <td>-83</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2035500</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>178</td>\n",
+                            "      <td>-51</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>178 rows × 4 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "         sample_index  time_index  eeg  y\n",
+                            "0                   0           1  135  0\n",
+                            "11500               0           2  190  0\n",
+                            "23000               0           3  229  0\n",
+                            "34500               0           4  223  0\n",
+                            "46000               0           5  192  0\n",
+                            "...               ...         ...  ... ..\n",
+                            "1989500             0         174 -103  0\n",
+                            "2001000             0         175 -127  0\n",
+                            "2012500             0         176 -116  0\n",
+                            "2024000             0         177  -83  0\n",
+                            "2035500             0         178  -51  0\n",
+                            "\n",
+                            "[178 rows x 4 columns]"
+                        ]
+                    },
+                    "execution_count": 13,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "data_unpivoted[data_unpivoted[\"sample_index\"] == 0]"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We then can have a look at some of the EEG signals and get a feel for what we are dealing with. We pick the first `n` (we chose 5, use any number you like) samples of every class and plot the EEG signals side-by-side. "
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 14,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "n = 5\n",
+                "\n",
+                "index_n_epileptic = data_unpivoted[data_unpivoted[\"y\"] == 1][\"sample_index\"].unique()[\n",
+                "    :n\n",
+                "]\n",
+                "index_n_nonepileptic = data_unpivoted[data_unpivoted[\"y\"] == 0][\n",
+                "    \"sample_index\"\n",
+                "].unique()[:n]\n",
+                "\n",
+                "samples_to_show = np.concatenate((index_n_epileptic, index_n_nonepileptic))"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 15,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 1106.61x500 with 2 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "g = sns.relplot(\n",
+                "    data=data_unpivoted[data_unpivoted[\"sample_index\"].isin(samples_to_show)],\n",
+                "    kind=\"line\",\n",
+                "    x=\"time_index\",\n",
+                "    y=\"eeg\",\n",
+                "    col=\"y\",\n",
+                "    hue=\"sample_index\",\n",
+                "    legend=\"full\",\n",
+                "    palette=sns.color_palette(),\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We can already guess that epileptic signals seem to be a lot more deviating than non-epileptic signals. Let's have a look at the standard deviation of EEG values per class label and compare them:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 16,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/plain": [
+                            "<seaborn.axisgrid.FacetGrid at 0x7969f37261a0>"
+                        ]
+                    },
+                    "execution_count": 16,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                },
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 511.111x500 with 1 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "g = sns.catplot(\n",
+                "    data=data_unpivoted.groupby([\"sample_index\", \"y\"]).std().reset_index(),\n",
+                "    kind=\"box\",\n",
+                "    x=\"y\",\n",
+                "    y=\"eeg\",\n",
+                ")\n",
+                "g.set_ylabels(\"std of eeg\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 17,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/plain": [
+                            "<seaborn.axisgrid.FacetGrid at 0x7969f238c550>"
+                        ]
+                    },
+                    "execution_count": 17,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                },
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 558.875x500 with 1 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "sns.displot(\n",
+                "    data=data_unpivoted.groupby([\"sample_index\", \"y\"]).std().reset_index(),\n",
+                "    kind=\"kde\",\n",
+                "    x=\"eeg\",\n",
+                "    hue=\"y\",\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "At this point, it is fairly obvious that standard deviation is going to be a major feature in this analysis."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Creating a GetML Data Model\n",
+                "\n",
+                "Now that we have explored our data, let's do some machine learning. GetML uses a highly sophisticated engine that runs in the background and takes away a lot of hassle in machine learning applications. \n",
+                "\n",
+                "Let's take a look at loading data into your getML project. First, let's learn how we work with data in getML. Data is represented by getML's custom [DataFrame](https://docs.getml.com/latest/api/data/getml.DataFrame.html) that behaves similarly to a pandas DataFrame. However, a [getML.DataFrame](https://docs.getml.com/latest/api/data/getml.DataFrame.html) is a representation of our data inside getML's highly efficient C++ database engine that runs in the background. We can [load data](https://docs.getml.com/latest/user_guide/importing_data/importing_data.html) from various sources such as pandas DataFrames ([`getml.DataFrame.from_pandas`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_pandas.html)), from CSV files ([`getml.DataFrame.from_csv`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_csv.html)), or load from remote databases ([`getml.DataFrame.from_db`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_db.html)) or even S3 buckets ([`getml.DataFrame.from_s3`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_s3.html)).\n",
+                "\n",
+                "Let's create a population DataFrame that contains our main goal: classify a 1s window. This means that we only need a DataFrame that holds the class labels of each window and a unique id, which in this case can just be the `sample_index`."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### The getML DataFrames"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 18,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "population_df = pd.DataFrame(\n",
+                "    {\n",
+                "        \"sample_index\": data.index.values,\n",
+                "        \"y\": data.y,\n",
+                "    }\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 19,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<div>\n",
+                            "<style scoped>\n",
+                            "    .dataframe tbody tr th:only-of-type {\n",
+                            "        vertical-align: middle;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe tbody tr th {\n",
+                            "        vertical-align: top;\n",
+                            "    }\n",
+                            "\n",
+                            "    .dataframe thead th {\n",
+                            "        text-align: right;\n",
+                            "    }\n",
+                            "</style>\n",
+                            "<table border=\"1\" class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr style=\"text-align: right;\">\n",
+                            "      <th></th>\n",
+                            "      <th>sample_index</th>\n",
+                            "      <th>y</th>\n",
+                            "    </tr>\n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    <tr>\n",
+                            "      <th>0</th>\n",
+                            "      <td>0</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>1</th>\n",
+                            "      <td>1</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>2</th>\n",
+                            "      <td>2</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>3</th>\n",
+                            "      <td>3</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>4</th>\n",
+                            "      <td>4</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>...</th>\n",
+                            "      <td>...</td>\n",
+                            "      <td>...</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11495</th>\n",
+                            "      <td>11495</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11496</th>\n",
+                            "      <td>11496</td>\n",
+                            "      <td>1</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11497</th>\n",
+                            "      <td>11497</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11498</th>\n",
+                            "      <td>11498</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "    <tr>\n",
+                            "      <th>11499</th>\n",
+                            "      <td>11499</td>\n",
+                            "      <td>0</td>\n",
+                            "    </tr>\n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "<p>11500 rows × 2 columns</p>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "       sample_index  y\n",
+                            "0                 0  0\n",
+                            "1                 1  1\n",
+                            "2                 2  0\n",
+                            "3                 3  0\n",
+                            "4                 4  0\n",
+                            "...             ... ..\n",
+                            "11495         11495  0\n",
+                            "11496         11496  1\n",
+                            "11497         11497  0\n",
+                            "11498         11498  0\n",
+                            "11499         11499  0\n",
+                            "\n",
+                            "[11500 rows x 2 columns]"
+                        ]
+                    },
+                    "execution_count": 19,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "population_df"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "A getML.DataFrame can be created from a pandas DataFrame as follows. As this getML.DataFrame is a representation of data internally handled by the engine, we need to specify an internal name. DataFrames are represented by these names in the monitor."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 20,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "population = getml.DataFrame.from_pandas(population_df, name=\"population\")"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 21,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th> name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"unused_float\">sample_index</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"unused_float\">           y</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">unused_float</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">unused_float</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">3</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11495</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">11495</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11496</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">11496</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11497</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">11497</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11498</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">11498</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11499</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">11499</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    11500 rows x 2 columns<br />\n",
+                            "    memory usage: 0.18 MB<br />\n",
+                            "    name: population<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            " name   sample_index              y\n",
+                            " role   unused_float   unused_float\n",
+                            "    0              0              0\n",
+                            "    1              1              1\n",
+                            "    2              2              0\n",
+                            "    3              3              0\n",
+                            "    4              4              0\n",
+                            "                 ...            ...\n",
+                            "11495          11495              0\n",
+                            "11496          11496              1\n",
+                            "11497          11497              0\n",
+                            "11498          11498              0\n",
+                            "11499          11499              0\n",
+                            "\n",
+                            "\n",
+                            "11500 rows x 2 columns\n",
+                            "memory usage: 0.18 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 21,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "population"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "As you can see, our data is now stored inside the engine and represented by a getML.DataFrame (data is of course the same). The Python API provides a link to the getML.DataFrame in the monitor, where you can conveniently explore your data.\n",
+                "\n",
+                "Now we need to [annotate our data](https://docs.getml.com/latest/user_guide/annotating_data/annotating_data.html) so the engine knows what to do with it.\n",
+                "\n",
+                "A key aspect of using getML.DataFrame are [roles](https://docs.getml.com/latest/api/getml.data.Roles.html). Every column with relevant data to our data model needs to have a certain role specified. As you can see, both of our columns have the `unused_float` role for now. One of the most important roles is [`getml.data.roles.target`](https://docs.getml.com/latest/api/roles/getml.data.roles.target.html), specifying that the data in this column is our target variable, the value that we want to train our machine learning model on. In our case, the column `y` containing the class label is our target. Let's tell the engine exactly that:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 22,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "population.set_role([\"y\"], getml.data.roles.target)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "As you may have noticed, our population getML.DataFrame does not contain any actual data, specifically no EEG signals. We utilize one of getML's core strengths here: relational data and time-series.\n",
+                "\n",
+                "Our data can in fact be interpreted as a relational time-series. We have a label for each window with a unique window id (`sample_index`) that stands in relation to its actual data, the corresponding EEG signal of each individual window. Each window has the EEG signal values along with the unique window id (`sample_index`). In other words: we can utilize a very efficient data model by joining a peripheral table containing the EEG values onto the window labels.\n",
+                "\n",
+                "Thus, the next step is to specify the `sample_index` as the join key using `getml.data.roles.join_key` in our population table:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 23,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "population.set_role([\"sample_index\"], getml.data.roles.join_key)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 24,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th> name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"join_key\">sample_index</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"target\">     y</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\"> role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header join_key\">    join_key</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header target\">target</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">1</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">2</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">3</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">4</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11495</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11495</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11496</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11496</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11497</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11497</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11498</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11498</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>11499</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align target\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    11500 rows x 2 columns<br />\n",
+                            "    memory usage: 0.14 MB<br />\n",
+                            "    name: population<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            " name   sample_index        y\n",
+                            " role       join_key   target\n",
+                            "    0              0        0\n",
+                            "    1              1        1\n",
+                            "    2              2        0\n",
+                            "    3              3        0\n",
+                            "    4              4        0\n",
+                            "                 ...      ...\n",
+                            "11495          11495        0\n",
+                            "11496          11496        1\n",
+                            "11497          11497        0\n",
+                            "11498          11498        0\n",
+                            "11499          11499        0\n",
+                            "\n",
+                            "\n",
+                            "11500 rows x 2 columns\n",
+                            "memory usage: 0.14 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 24,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "population"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We now create the peripheral table containing the time-series data corresponding to the 1s window labels in the population table just like we did previously with our population table:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 25,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "peripheral = getml.DataFrame.from_pandas(data_unpivoted, name=\"peripheral\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now we need to specify the column roles in the peripheral table. This getML.DataFrame contains the `sample_index` as well and we need to set it as our join key, as described above. Subsequently, as this table will contain our actual data in the form of EEG signal values, we specify the role of this column as numerical ([`getml.data.roles.numerical`](https://docs.getml.com/latest/api/roles/getml.data.roles.numerical.html)), something we can train our machine learning model on:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 26,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "peripheral.set_role([\"sample_index\"], getml.data.roles.join_key)\n",
+                "peripheral.set_role([\"eeg\"], getml.data.roles.numerical)"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 27,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  table {\n",
+                            "    font-family: Helvetica, sans-serif;\n",
+                            "  }\n",
+                            "\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right !important;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.sub-header {\n",
+                            "    font-weight: normal;\n",
+                            "    font-style: italic;\n",
+                            "  }\n",
+                            "  .join_key,\n",
+                            "  .numerical,\n",
+                            "  .target,\n",
+                            "  .unused_float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "\n",
+                            "  .char-align {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  span.left {\n",
+                            "    text-align: right;\n",
+                            "    width: 3em;\n",
+                            "  }\n",
+                            "  span.right {\n",
+                            "    float: right;\n",
+                            "    text-align: left;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th>   name</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"join_key\">sample_index</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"numerical\">      eeg</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"unused_float\">  time_index</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"unused_float\">           y</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header\">   role</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header join_key\">    join_key</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header numerical\">numerical</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">unused_float</th>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            <th class=\"sub-header unused_float\">unused_float</th>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">135</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">1</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">190</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">229</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">3</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">223</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">0</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">192</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">5</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">...</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">...</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2046995</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">5</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">174</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2046996</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">4</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">175</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2046997</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">-2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">176</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2046998</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">2</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">177</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2046999</th>\n",
+                            "        \n",
+                            "          \n",
+                            "            <td class=\"join_key\">11499</td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align numerical\">\n",
+                            "              <span class=\"left\">20</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">178</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "          \n",
+                            "            \n",
+                            "              \n",
+                            "              \n",
+                            "          \n",
+                            "            <td class=\"char-align unused_float\">\n",
+                            "              <span class=\"left\">0</span\n",
+                            "              ><span class=\"right\" style=\"width: 0ch\"\n",
+                            "                >&nbsp;</span\n",
+                            "              >\n",
+                            "            </td>\n",
+                            "          \n",
+                            "        \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "\n",
+                            "  <p>\n",
+                            "    2047000 rows x 4 columns<br />\n",
+                            "    memory usage: 57.32 MB<br />\n",
+                            "    name: peripheral<br />\n",
+                            "    type: getml.DataFrame<br />\n",
+                            "    \n",
+                            "  </p>\n"
+                        ],
+                        "text/plain": [
+                            "   name   sample_index         eeg     time_index              y\n",
+                            "   role       join_key   numerical   unused_float   unused_float\n",
+                            "      0              0        135               1              0\n",
+                            "      1              0        190               2              0\n",
+                            "      2              0        229               3              0\n",
+                            "      3              0        223               4              0\n",
+                            "      4              0        192               5              0\n",
+                            "                   ...         ...            ...            ...\n",
+                            "2046995          11499          5             174              0\n",
+                            "2046996          11499          4             175              0\n",
+                            "2046997          11499         -2             176              0\n",
+                            "2046998          11499          2             177              0\n",
+                            "2046999          11499         20             178              0\n",
+                            "\n",
+                            "\n",
+                            "2047000 rows x 4 columns\n",
+                            "memory usage: 57.32 MB\n",
+                            "type: getml.DataFrame"
+                        ]
+                    },
+                    "execution_count": 27,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "peripheral"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "As you may have noticed, there are still `unused_float` columns left. This data is present, but we do not use or need it in our machine learning efforts. This unused data is not considered and can just be ignored."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### The getML Data Model\n",
+                "\n",
+                "Now that we have our data efficiently stored in getML.DataFrame, we continue to construct our data model.\n",
+                "\n",
+                "This is very easily done by using one of getML's many [DataModels](https://docs.getml.com/latest/user_guide/data_model/data_model.html). We put our time-series data in a relational context and can utilze for example a simple [StarSchema](https://docs.getml.com/latest/api/getml.data.StarSchema.html) data model to accomplish this. Easily put, we see our windows (the time-series data) as splits into many individual samples that are joined onto the window labels. This way, we are effectively thinking of time series as relational data: we are identifying relevant information from our data and aggragate it into a single label. In fact, what we are doing is effectively a self join, because we are joining a table to itself. This allows for very efficient calculation.\n",
+                "\n",
+                "\n",
+                "First, we define a random data [split](https://docs.getml.com/latest/api/getml.data.split.html):"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 28,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "train, test = 0.9, 0.1\n",
+                "\n",
+                "split = getml.data.split.random(seed=5849, train=train, test=test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Second, we create our data model. We create a StarSchema containing our population getML.DataFrame as the population table and specify the split of our dataset into train and test set. We then [join](https://docs.getml.com/latest/api/StarSchema/getml.data.StarSchema.join.html) our peripheral table to our time series on the join key, in this case `sample_index`:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 29,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "time_series = getml.data.StarSchema(population=population, split=split)\n",
+                "\n",
+                "time_series.join(\n",
+                "    peripheral,\n",
+                "    on=\"sample_index\",\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "getML provides a convenient view to our data model. We can look at a diagram representation of our data model with table names and specific joins, as well as the staging tables and statistics about the underlying data container."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 30,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
+                            "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
+                            "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">peripheral</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"45.0\" x=\"324.0\" font-size=\"7pt\" >sample_index = sample_index</tspan></text></svg></div>\n",
+                            "            </div>\n",
+                            "\n",
+                            "            <div style='margin-top: 15px;'>\n",
+                            "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
+                            "            <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">data frames</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">staging table              </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">peripheral</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">PERIPHERAL__STAGING_TABLE_2</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "            </div>\n",
+                            "            \n",
+                            "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
+                            "<div style='margin-top: 15px;'>\n",
+                            "<div style='float: left; margin-right: 50px;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">subset</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name      </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> rows</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">1086</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">population</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">10414</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">View</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "<div style='float: left;'>\n",
+                            "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
+                            "    <style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name      </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\">   rows</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">type     </th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">peripheral</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"int\">2047000</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">DataFrame</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>\n",
+                            "</div>\n",
+                            "</div>"
+                        ],
+                        "text/plain": [
+                            "data model\n",
+                            "\n",
+                            "  population:\n",
+                            "    columns:\n",
+                            "    - sample_index: join_key\n",
+                            "    - y: target\n",
+                            "\n",
+                            "    joins:\n",
+                            "    - right: 'peripheral'\n",
+                            "      on: (population.sample_index, peripheral.sample_index)\n",
+                            "      relationship: 'many-to-many'\n",
+                            "      lagged_targets: False\n",
+                            "\n",
+                            "  peripheral:\n",
+                            "    columns:\n",
+                            "    - sample_index: join_key\n",
+                            "    - eeg: numerical\n",
+                            "    - time_index: unused_float\n",
+                            "    - y: unused_float\n",
+                            "\n",
+                            "\n",
+                            "container\n",
+                            "\n",
+                            "  population\n",
+                            "      subset   name          rows   type\n",
+                            "  0   test     population    1086   View\n",
+                            "  1   train    population   10414   View\n",
+                            "\n",
+                            "  peripheral\n",
+                            "      name            rows   type     \n",
+                            "  0   peripheral   2047000   DataFrame"
+                        ]
+                    },
+                    "execution_count": 30,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "time_series"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "This is an overview of your data model in the getML engine. At the top you can see a visual representation in the form of a diagram. Here you can easily see how your data and the specific joins is structured. Next you are presented the so called staging tables. This is a list of the relevant data frames and staging table names. At last, you can see an overview of all the data [containers](https://docs.getml.com/latest/api/getml.data.Container.html). This includes the split in train and test set of your population table as well as the peripheral tables.\n",
+                "\n",
+                "In this simple example, the diagram consists of a single join of the peripheral table onto the population table via the `sample_index` as a join key. The population table is split into 90% train and 10% test set. The peripheral talbe contains all the EEG signal values and has over 2 million rows."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "### The getML machine learning pipeline\n",
+                "\n",
+                "Complex machine learning models are represented by getML [pipelines](https://docs.getml.com/latest/api/pipeline/getml.pipeline.Pipeline.html). A pipeline contains the data model (including complex data relations), data [preprocessors](https://docs.getml.com/latest/api_reference/preprocessors.html), [feature learners](https://docs.getml.com/latest/api_reference/feature_learning.html), [predictors](https://docs.getml.com/latest/api_reference/predictors.html) and so on.\n",
+                "\n",
+                "In our approach, we will use getML's very own [FastProp](https://docs.getml.com/latest/api/getml.feature_learning.FastProp.html) automatic feature learner for [feature engineering](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html). We specify a loss function suitable for classification. As we are only dealing with a univariate time-series, we want to use all possible aggregation functions.\n",
+                "\n",
+                "We use the highly efficient [XGBoost](https://docs.getml.com/latest/api/getml.predictors.XGBoostClassifier.html) classifier algorithm as a [predictor](https://docs.getml.com/latest/user_guide/predicting/predicting.html)."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 31,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "feature_learner = getml.feature_learning.FastProp(\n",
+                "    loss_function=getml.feature_learning.loss_functions.CrossEntropyLoss,\n",
+                "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
+                ")\n",
+                "\n",
+                "predictor = getml.predictors.XGBoostClassifier()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now that we have our data model and machine learning components defined in just a few lines of code, we declare our machine learning pipeline as simple as the following. For convenience, we specify some free to choose tags. These are shown in the monitor and can be used to efficiently and easily distinguish different pipelines and their performance."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 32,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "pipe = getml.pipeline.Pipeline(\n",
+                "    data_model=time_series.data_model,\n",
+                "    tags=[\"FastProp+AggAll\", \"XGBoost\", f\"split={train}/{test}\"],\n",
+                "    feature_learners=[feature_learner],\n",
+                "    predictors=[predictor],\n",
+                ")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now all we need to do is train our model:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 33,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "FastProp: Trying 25 features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
+                        "\n",
+                        "Trained pipeline.\n",
+                        "Time taken: 0h:0m:1.434615\n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<pre>Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='CrossEntropyLoss',\n",
+                            "         peripheral=['peripheral'],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['FastProp+AggAll', 'XGBoost', 'split=0.9/0.1', 'container-Mq37qh'])</pre>"
+                        ],
+                        "text/plain": [
+                            "Pipeline(data_model='population',\n",
+                            "         feature_learners=['FastProp'],\n",
+                            "         feature_selectors=[],\n",
+                            "         include_categorical=False,\n",
+                            "         loss_function='CrossEntropyLoss',\n",
+                            "         peripheral=['peripheral'],\n",
+                            "         predictors=['XGBoostClassifier'],\n",
+                            "         preprocessors=[],\n",
+                            "         share_selected_features=0.5,\n",
+                            "         tags=['FastProp+AggAll', 'XGBoost', 'split=0.9/0.1', 'container-Mq37qh'])"
+                        ]
+                    },
+                    "execution_count": 33,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe.fit(time_series.train, check=False)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "This is all it takes. FastProp found features in just 1 second and XGBoost trained our model in just 2 seconds. The whole process took less than 4 seconds!"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Model Evaluation"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Now, let's look at how well our model performs. Again, getML does everything for you. We [score](https://docs.getml.com/latest/api/pipeline/Pipeline/getml.pipeline.Pipeline.score.html) our pipeline on the test set:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 34,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
+                        "\n"
+                    ]
+                },
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\"> </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"datetime\">date time          </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">set used</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">accuracy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">    auc</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">cross entropy</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-08-26 15:14:26</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">train</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.98</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9974</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.05586</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"datetime\">2024-08-26 15:14:26</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">test</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.9696</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.994</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0793</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "    date time             set used   target   accuracy       auc   cross entropy\n",
+                            "0   2024-08-26 15:14:26   train      y          0.98      0.9974         0.05586\n",
+                            "1   2024-08-26 15:14:26   test       y          0.9696    0.994          0.0793 "
+                        ]
+                    },
+                    "execution_count": 34,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe.score(time_series.test)"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Let's have a look at some key [machine learning metrics](https://docs.getml.com/latest/api/pipeline/getml.pipeline.Scores.html): Accuracy and Area Under Curve (AUC):"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 35,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "Accuracy: 96.96%, AUC: 0.9940\n"
+                    ]
+                }
+            ],
+            "source": [
+                "print(f\"Accuracy: {pipe.scores.accuracy*100:.2f}%, AUC: {pipe.scores.auc:.4f}\")"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "There are several other, more complex metrics to understand the performance of a machine learning model. The most prominent being Receiver Operating Characteristic (ROC) curve, precision-recall curve and lift curve. getML has these already calculated for you:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 36,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 1200x400 with 3 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "# Refers to the data from the last time we called .score(...)\n",
+                "fpr, tpr = pipe.plots.roc_curve()\n",
+                "recall, precision = pipe.plots.precision_recall_curve()\n",
+                "proportion, lift = pipe.plots.lift_curve()\n",
+                "\n",
+                "fig, ax = plt.subplots(ncols=3, figsize=(12, 4))\n",
+                "\n",
+                "ax[0].plot(fpr, tpr, color=\"#6829c2\")\n",
+                "ax[0].set_title(\"receiver operating characteristic (ROC)\")\n",
+                "ax[0].set_xlabel(\"false positive rate\")\n",
+                "ax[0].set_ylabel(\"true positive rate\")\n",
+                "\n",
+                "ax[1].plot(recall, precision, color=\"#6829c2\")\n",
+                "ax[1].set_title(\"precision-recall curve\")\n",
+                "ax[1].set_xlabel(\"recall (true positive rate)\")\n",
+                "ax[1].set_ylabel(\"precision\")\n",
+                "\n",
+                "ax[2].plot(proportion, lift, color=\"#6829c2\")\n",
+                "ax[2].set_title(\"lift curve\")\n",
+                "ax[2].set_xlabel(\"proportion\")\n",
+                "ax[2].set_ylabel(\"lift\")\n",
+                "\n",
+                "plt.show()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "The Receiver-Operator-Characteristic (ROC) curve is a diagram that shows the diagnostic ability of a binary classifier for all classification thresholds. It shows the tradeoff between sensitivity (true positive rate or TPR) and specificity (1 - FPR or false positive rate). Easily put, the closer the curve to the top left (meaning the larger the area under curve or AUC) , the more accurate the classifier. A 45° diagonal would be a random classifier.\n",
+                "\n",
+                "Much like the ROC curve, a precision-recall curve (PR curve) is used to evaluate the performance of a binary classifier. It is often used when dealing with heavily imbalanced classes. It is desired that your machine learning model has both high precision and high recall. However, we often end up with a trade-off between the two. Similar to a ROC curve, the higher the area und the curve the better performing our binary classifier.\n",
+                "\n",
+                "The Lift curve shows the relation between the number of instances which were predicted positive and those that are indeed positive and thus, like ROC and PR curves, measures the effectiveness of a chosen classifier against a random classifier. In our example, the patients with the highest probability of having an epileptic seizure appear on the left of the Lift curve along with high Lift scores. This point is called the Maximum Lift Point: the higher this point, the better our model performs. Also, it is generally considered that the longer the flat part on the right of the Lift curve, the more reliable the model is.\n",
+                "\n",
+                "All three performance diagrams measure the performance of a binary classifier against a random classifier. As a rule of thumb, the higher ROC and PR curves the better, while a Lift curve is desired to be high in the left and preferably flat on the right."
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Studying the Features\n",
+                "\n",
+                "Finally, we can have a look at the features our relational learning algorithm has extracted. We can view them conveniently in the getML monitor under the respective pipeline, or print them directly in Python:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 37,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/html": [
+                            "<style>\n",
+                            "  th {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  td {\n",
+                            "    text-align: left !important;\n",
+                            "  }\n",
+                            "  th:nth-child(1) {\n",
+                            "    text-align: right;\n",
+                            "    border-right: 1px solid LightGray;\n",
+                            "  }\n",
+                            "  th.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.float {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  th.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "  td.int {\n",
+                            "    text-align: right !important;\n",
+                            "  }\n",
+                            "</style>\n",
+                            "\n",
+                            "<table class=\"dataframe\">\n",
+                            "  <thead>\n",
+                            "    <tr>\n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"int\">  </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">target</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"str\">name        </th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">correlation</th>\n",
+                            "        \n",
+                            "      \n",
+                            "        \n",
+                            "          <th class=\"float\">importance</th>\n",
+                            "        \n",
+                            "      \n",
+                            "    </tr>\n",
+                            "    \n",
+                            "  </thead>\n",
+                            "  <tbody>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>0</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_1</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0358</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0087</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>1</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_2</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.7295</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.018</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>2</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_3</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">-0.7295</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>3</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_4</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.7646</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.008</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>4</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_5</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0516</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0041</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th></th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">...</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">...</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">...</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">...</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>20</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_21</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.7634</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0012</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>21</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_22</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.7615</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0017</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>22</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_23</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0186</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0189</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>23</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_24</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">-0.0433</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0026</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "      <tr>\n",
+                            "        <th>24</th>\n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">y</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"str\">feature_1_25</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0</td>\n",
+                            "            \n",
+                            "          \n",
+                            "            \n",
+                            "              <td class=\"float\">0.0</td>\n",
+                            "            \n",
+                            "          \n",
+                            "      </tr>\n",
+                            "    \n",
+                            "  </tbody>\n",
+                            "</table>"
+                        ],
+                        "text/plain": [
+                            "     target   name           correlation   importance\n",
+                            " 0   y        feature_1_1         0.0358       0.0087\n",
+                            " 1   y        feature_1_2         0.7295       0.018 \n",
+                            " 2   y        feature_1_3        -0.7295       0.0   \n",
+                            " 3   y        feature_1_4         0.7646       0.008 \n",
+                            " 4   y        feature_1_5         0.0516       0.0041\n",
+                            "     ...      ...                    ...          ...\n",
+                            "20   y        feature_1_21        0.7634       0.0012\n",
+                            "21   y        feature_1_22        0.7615       0.0017\n",
+                            "22   y        feature_1_23        0.0186       0.0189\n",
+                            "23   y        feature_1_24       -0.0433       0.0026\n",
+                            "24   y        feature_1_25        0.0          0.0   "
+                        ]
+                    },
+                    "execution_count": 37,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe.features"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "We can look at feature correlations:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 38,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 800x400 with 1 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "names, correlations = pipe.features.correlations()\n",
+                "\n",
+                "plt.subplots(figsize=(8, 4))\n",
+                "\n",
+                "plt.bar(names, correlations, color=\"#6829c2\")\n",
+                "\n",
+                "plt.title(\"feature correlations\")\n",
+                "plt.xlabel(\"features\")\n",
+                "plt.ylabel(\"correlations\")\n",
+                "plt.xticks(rotation=\"vertical\")\n",
+                "\n",
+                "plt.show()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "Or we can have a look at the feature importances. This is particularly interesting if we want to understand what generated features are most important for our machine learning model.\n",
+                "\n",
+                "The feature importance is calculated by XGBoost based on the improvement of the optimizing criterium at each split in the decision tree and is normalized to 100%."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 39,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "image/png": "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",
+                        "text/plain": [
+                            "<Figure size 800x400 with 1 Axes>"
+                        ]
+                    },
+                    "metadata": {},
+                    "output_type": "display_data"
+                }
+            ],
+            "source": [
+                "names, importances = pipe.features.importances()\n",
+                "\n",
+                "plt.subplots(figsize=(8, 4))\n",
+                "\n",
+                "plt.bar(names, importances, color=\"#6829c2\")\n",
+                "\n",
+                "plt.title(\"feature importances\")\n",
+                "plt.xlabel(\"features\")\n",
+                "plt.ylabel(\"importance\")\n",
+                "plt.xticks(rotation=\"vertical\")\n",
+                "\n",
+                "plt.show()"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "This is intriguing! It seems that one particular feature stands out in its importance for classification predictions.\n",
+                "\n",
+                "As we already discussed above, we view our time series data from a relation data point of view. In fact, relational learning is one of getML's core strengths. Particularly, getML is able to transpile features into database queries that can be used in production database environments without the need of any other software component.\n",
+                "\n",
+                "Let's have a look at the SQL code of our most important feature:"
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": 40,
+            "metadata": {},
+            "outputs": [
+                {
+                    "data": {
+                        "text/markdown": [
+                            "```sql\n",
+                            "DROP TABLE IF EXISTS \"FEATURE_1_7\";\n",
+                            "\n",
+                            "CREATE TABLE \"FEATURE_1_7\" AS\n",
+                            "SELECT STDDEV( t2.\"eeg\" ) AS \"feature_1_7\",\n",
+                            "       t1.rowid AS rownum\n",
+                            "FROM \"POPULATION__STAGING_TABLE_1\" t1\n",
+                            "INNER JOIN \"PERIPHERAL__STAGING_TABLE_2\" t2\n",
+                            "ON t1.\"sample_index\" = t2.\"sample_index\"\n",
+                            "GROUP BY t1.rowid;\n",
+                            "```"
+                        ],
+                        "text/plain": [
+                            "'DROP TABLE IF EXISTS \"FEATURE_1_7\";\\n\\nCREATE TABLE \"FEATURE_1_7\" AS\\nSELECT STDDEV( t2.\"eeg\" ) AS \"feature_1_7\",\\n       t1.rowid AS rownum\\nFROM \"POPULATION__STAGING_TABLE_1\" t1\\nINNER JOIN \"PERIPHERAL__STAGING_TABLE_2\" t2\\nON t1.\"sample_index\" = t2.\"sample_index\"\\nGROUP BY t1.rowid;'"
+                        ]
+                    },
+                    "execution_count": 40,
+                    "metadata": {},
+                    "output_type": "execute_result"
+                }
+            ],
+            "source": [
+                "pipe.features.to_sql()[names[0]]"
+            ]
+        },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "As you can see, the most important feature in seizure recognition from EEG signals seems to be the standard deviation. Just like we guessed in the beginning of this notebook.\n",
+                "\n",
+                "However, relevant features are by far not always so obvious as in this particular example dataset. In fact, most of the time feature engineering takes a lot of effort and domain knowledge from domain experts. As we discussed above, manual feature engineering is a cumbersome, time consuming and error prone process.\n",
+                "\n",
+                "Novel machine learning libraries like getML with automatic feature learning, flexible data models and machine learning pipelines, all wrapped inside an easy to use Python API, backed by an efficient and fast C++ backend make this task a lot easier and way more efficient for data scientists. "
+            ]
+        }
+    ],
+    "metadata": {
+        "kernelspec": {
+            "display_name": "Python 3 (ipykernel)",
+            "language": "python",
+            "name": "python3"
+        },
+        "language_info": {
+            "codemirror_mode": {
+                "name": "ipython",
+                "version": 3
+            },
+            "file_extension": ".py",
+            "mimetype": "text/x-python",
+            "name": "python",
+            "nbconvert_exporter": "python",
+            "pygments_lexer": "ipython3",
+            "version": "3.10.12"
+        }
     },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 511.111x500 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "g = sns.catplot(\n",
-    "    data=data_unpivoted.groupby([\"sample_index\", \"y\"]).std().reset_index(),\n",
-    "    kind=\"box\",\n",
-    "    x=\"y\",\n",
-    "    y=\"eeg\",\n",
-    ")\n",
-    "g.set_ylabels(\"std of eeg\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "<seaborn.axisgrid.FacetGrid at 0x7969f238c550>"
-      ]
-     },
-     "execution_count": 17,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiwAAAHpCAYAAAChumdzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABhYElEQVR4nO3deXhTZeI+/Dtrd0ppCwJFERhaKF1xBukUGAsyLDJfdmcpDAUEpag4KKA4SnEpOugAiopssjn+GFlel+IgIozYMi60UBBBZJcC6UZbumQ77x9pThuStkma5ITm/lxXLtqTJzlPHijcPKtMEAQBRERERF5MLnUFiIiIiFrCwEJERERej4GFiIiIvB4DCxEREXk9BhYiIiLyegwsRERE5PUYWIiIiMjrMbC4kSAIMBgM4FY3RERErcPA4kZGoxEFBQUwGo2S1+PYsWOS18PbsZ3sw3ZqGdvIPmwncgQDiw8QBAE6nY49PS1gO9mH7dQytpF92E7kCAYWIiIi8noMLEREROT1GFiIiIjI6zGwEBERkddjYCEiIiKvx8BCREREXo+BhYiIiLweAwsRERF5PQYWIiIi8nqSB5Zt27YhLS0NcXFxmDRpEo4dO9Zs+T179mDEiBGIi4vDmDFjcPDgQYvn9+7di+nTp2PAgAGIjo7GyZMnLZ6/fPkyoqOjbT727NkjlrP1/Keffuq6D05ERER2kzSw5OTkIDs7G5mZmdi1axdiYmIwY8YMlJSU2Cx/5MgRzJ8/HxMnTsTu3bsxdOhQZGZm4vTp02KZ6upqJCcn48knn7T5Hp07d8ahQ4csHo8++igCAwMxePBgi7LZ2dkW5YYNG+a6D09ERER2U0p5840bN2Ly5MmYMGECACArKwsHDhzAjh07MGvWLKvymzdvxqBBgzBz5kwAwLx585Cbm4utW7di6dKlAICxY8cCMPWk2KJQKBAZGWlxbd++fRg5ciSCgoIsrrdr186qLBEREXmeZIFFq9XixIkTmD17tnhNLpcjJSUF+fn5Nl9TUFCAadOmWVxLTU3Fvn37nK7H8ePHcfLkSTz33HNWz2VlZWHx4sXo1q0b/vjHP2LChAmQyWQO38NgMDhdP1cw31/qeng7tpN92E4tYxvZx1faSaFQSF2FNkGywFJWVgaDwYDw8HCL6+Hh4Th79qzN1xQXFyMiIsKqfHFxsdP1+PDDD9GzZ08kJydbXH/sscdw7733IiAgAIcOHUJWVhaqq6sxdepUh+9RWFjodP1cyVvq4e3YTvZhO7WMbWSftt5O/fv3l7oKbYKkQ0JSq62txSeffII5c+ZYPZeZmSl+3bdvX9TU1GD9+vVOBZa4uDhJE7bBYEBhYaHk9fB2bCf7sJ1axjayD9uJHCFZYAkLC4NCobCaYFtSUmLVi2IWERFh1ZvSXPmWfPbZZ6itrRXnvTQnISEBb731FrRaLdRqtUP3USgUXvHD6C318HZsJ/uwnVrGNrIP24nsIdkqIbVajdjYWOTl5YnXjEYj8vLykJSUZPM1iYmJOHz4sMW13NxcJCYmOlWHHTt2IC0tDR06dGix7MmTJxEaGupwWCEiIqLWk3RIKCMjAwsXLkS/fv0QHx+PTZs2oaamBuPHjwcALFiwAJ06dcL8+fMBAFOnTsWUKVOwYcMGDBkyBDk5OTh+/Li4QggAysvLUVRUhOvXrwMAzp07B8DUO9N4xc+FCxfw7bff4t1337Wq1/79+1FSUoKEhAT4+fnh66+/xpo1azB9+nS3tYU3ulmnxwuf/IA/JHRBSi/nerGIiIhcQdLAMmrUKJSWlmLVqlXQaDTo06cP1q1bJw7xFBUVQS5v6ARKTk7G8uXLsWLFCrz++uvo3r07Vq9ejd69e4tl9u/fj6efflr8/oknngAAzJ07F48++qh4fceOHbjjjjuQmppqVS+lUolt27bh5ZdfBgDceeedWLRoESZPnuzaBvByO49cxgffXsIH317CsSXD0c5fJXWViIjIR8kEQRCkrkRbZTAYUFBQgMTERMkn3TpTj9f2nsIb+88AAB5N64X5w6PdVUWv4C2/X96O7dQytpF92E7kCMm35ifvVVxVJ3598LRGwpoQEZGvY2ChJhXdqBW/LqvWSlgTIiLydQws1KSrjQJL+U2dhDUhIiJfx8BCTWrcw1JZp4fOYJSwNkRE5MsYWMimaq0eN2ose1XKq9nLQkRE0mBgIZvMw0FBagXaB5qWM5dzHgsREUmEgYVsulphCix3hPojLNC0u28Ze1iIiEgiPn34ITXN3MNyR6g/qrWmo9+5UoiIiKTCwEI2mXtTwgLV8FOaAguHhIiISCoMLGTTzTo9ACDEXwm10jRyyCEhIiKSCgML2WQOLEFqJeqnsKDsJntYiIhIGgwsZFOVObD4KaFSyABwDgsREUmHgYVsqmo0JOSvMh1KxiEhIiKSCgML2XSzUQ9LaAD3YSEiImlxHxayqcpGYKmo0UtZJSIi8mEMLGTTzTrTUuZgPwUC1KYhoWodAwsREUmDgYVsMvewBPupEFA/h6VGy8MPiYhIGgwsZFPDkJACgWpzYGEPCxERSYOBhWy6KfawKBsNCRkgCIKU1SIiIh/FwEJWDEZBPD8oyE8pDgkJAlCn57AQERF5HgMLWbnZaOgnuFFgAYCa+iBDRETkSQwsZMU8HKSUy+CnlEOpkEOtMP1RqdExsBARkecxsJCVxpvGyWSmbfnFeSzsYSEiIgkwsJCVKnEPloaNkBuWNjOwEBGR5zGwkJXGK4TMxKXNHBIiIiIJMLCQlcrahj1YzBqGhLgXCxEReR4DC1lpPIfFzDwkVMseFiIikgADC1kxL2u2mMPCSbdERCQhBhayYj74MFBt3cPCwEJERFJgYCEr5mEff1XDHw/zpFsOCRERkRQYWMhKrd4cWBpPujX1trCHhYiIpMDAQlbqdKbzghr3sIj7sLCHhYiIJMDAQlbEISFlQw+LuA8Le1iIiEgCDCxkxXwis+WQEPdhISIi6TCwkBVbk24bhoSMktSJiIh8GwMLWTEHFj+VrSEh9rAQEZHnMbCQlVpdc0NCnMNCRESex8BCVsRlzUquEiIiIu/AwEJWmuth4SohIiKSAgMLWakzz2FRWu90yx4WIiKSAgMLWWlYJdTQw+JXvydLHVcJERGRBBhYyEqtjX1YzL0tdXr2sBARkecxsJAVW/uwmHtYtHr2sBARkecxsJAFQRBsDgmpxR4WBhYiIvI8BhayoDMIMAqmrxufJWQeEtIbBegNDC1ERORZDCxkobbRHBW/xkNCjb7WMrAQEZGHMbCQBfNwkExmuaxZrWj4miuFiIjI0xhYyII5jPgp5ZDJZOJ1pUIOpdz0PeexEBGRp0keWLZt24a0tDTExcVh0qRJOHbsWLPl9+zZgxEjRiAuLg5jxozBwYMHLZ7fu3cvpk+fjgEDBiA6OhonT560eo8pU6YgOjra4vHcc89ZlLly5QpmzZqFhIQEDBw4EK+88gr0+rZ/8J+tCbdm5h4XrhQiIiJPkzSw5OTkIDs7G5mZmdi1axdiYmIwY8YMlJSU2Cx/5MgRzJ8/HxMnTsTu3bsxdOhQZGZm4vTp02KZ6upqJCcn48knn2z23pMnT8ahQ4fEx4IFC8TnDAYDZs+eDZ1Ohw8++ADLli3Drl27sGrVKtd8cC8mbsuvtA4sau7FQkREElFKefONGzdi8uTJmDBhAgAgKysLBw4cwI4dOzBr1iyr8ps3b8agQYMwc+ZMAMC8efOQm5uLrVu3YunSpQCAsWPHAgAuX77c7L39/f0RGRlp87lDhw7hzJkz2LhxIyIiItCnTx88/vjjWL58OebOnQu1Wu3Q5zQYpP0H3nx/e+pRXacDYJpke2t5cw9LdZ1O8s/kDo60ky9jO7WMbWQfX2knhcL6P4DkOMkCi1arxYkTJzB79mzxmlwuR0pKCvLz822+pqCgANOmTbO4lpqain379jl8/48//hgfffQRIiMjcd9992HOnDkICAgQ79O7d29ERERY3GfJkiU4c+YM+vbt69C9CgsLHa6fO9hTjxPX6gAAgl6LgoICyycNpiGx4ydPQa9xLLTdTrzl98vbsZ1axjayT1tvp/79+0tdhTZBssBSVlYGg8GA8PBwi+vh4eE4e/aszdcUFxdbhAhz+eLiYofu/cADD6BLly7o2LEjTp06heXLl+PcuXN48803m7yP+XuNRuPQvQAgLi5O0oRtMBhQWFhoVz1KTl4HUIawkCAkJiZaPBdy8BCu3qzCnXf3RGLPcJuvv5050k6+jO3UMraRfdhO5AhJh4Sk8uCDD4pfR0dHIzIyEtOmTcPFixdx5513uvx+CoXCK34Y7amHtn7XOH+VdVnzRFy9sW13cXrL75e3Yzu1jG1kH7YT2UOySbdhYWFQKBRWE2xLSkqsejfMIiIirHpTmitvr4SEBADAhQsXmryP+fum5r20FeKk22ZWCXFZMxEReZpkgUWtViM2NhZ5eXniNaPRiLy8PCQlJdl8TWJiIg4fPmxxLTc312rowlHmpc/mMJKYmIjTp09bhKnc3FwEBwejV69erbqXt7N18KEZVwkREZFUJB0SysjIwMKFC9GvXz/Ex8dj06ZNqKmpwfjx4wEACxYsQKdOnTB//nwAwNSpUzFlyhRs2LABQ4YMQU5ODo4fPy6uEAKA8vJyFBUV4fr16wCAc+fOATD1mkRGRuLixYv4+OOPMWTIELRv3x6nTp1CdnY2fv3rXyMmJgaAaYJtr169sGDBAjz11FPQaDRYsWIF/vKXvzi8Quh2Y88+LOxhISIiT5M0sIwaNQqlpaVYtWoVNBoN+vTpg3Xr1olDPEVFRZDLG/6nn5ycjOXLl2PFihV4/fXX0b17d6xevRq9e/cWy+zfvx9PP/20+P0TTzwBAJg7dy4effRRqFQq5OXlYfPmzaiurkbnzp0xfPhwzJkzR3yNQqHAO++8gyVLluDBBx9EQEAAxo0bh8cee8zdTSI5cxixtQ+LX/01BhYiIvI0ySfdpqenIz093eZzW7Zssbo2cuRIjBw5ssn3Gz9+vNhDY0vnzp2xdevWFuvVtWtXrF27tsVybU1dfQ+Ln40hIfM1cxkiIiJPkXxrfvIudfUnMTc+7NCMQ0JERCQVBhayYD4nSK1sbtItAwsREXkWAwtZaC6wmOew8PBDIiLyNAYWstB8YOGyZiIikgYDC1nQNjuHhauEiIhIGgwsZMHcw+Jnq4dFXCXEwEJERJ7FwEIWOCRERETeiIGFLIhDQlwlREREXoSBhSyYw4jaxsmpXCVERERSYWAhCxwSIiIib8TAQhbsCyzsYSEiIs9iYCELzS5rrj/BmauEiIjI0xhYyEKzW/MrOCRERETSYGAhC3btw8IhISIi8jAGFrLQ3LJmc4jhKiEiIvI0BhayIA4JcWt+IiLyIgwsZIHLmomIyBsxsJBIEITmh4QazWERBMGjdSMiIt/GwEKixkM9NgNL/e63ggDoDAwsRETkOQwsJDL3rgBN7cPScI3DQkRE5EkMLCRqvPrHVmBpfI0rhYiIyJMYWEhkDiEqhQxyuczqeblc1mjzOAYWIiLyHAYWEjW3pNmM5wkREZEUGFhI1NwKIbOGlUKcw0JERJ7DwEKi5vZgMROHhHgAIhEReRADC4nq7Ags4onNHBIiIiIPYmAhkSNzWLhKiIiIPImBhUQNc1gUTZbh9vxERCQFBhYS2TOHhQcgEhGRFBhYSGQOLH7NDAmp2cNCREQSYGAhkdZgCiHN97BwlRAREXkeAwuJ7BoSUnHjOCIi8jwGFhLZt0pIYVGWiIjIExhYSGTuNWl8KvOtuEqIiIikwMBCInFZM88SIiIiL8PAQiK7tuZnYCEiIgkwsJDIoX1YdBwSIiIiz2FgIZGOQ0JEROSlGFhIpDMIAABVc4FFxbOEiIjI8xhYSGTuYVEqZE2W4db8REQkBQYWEunt6GHh1vxERCQFBhYSmXtYVM32sHAOCxEReR4DC4l0RlMPi1JuzyohBhYiIvIcBhYS6c09LPYcfsghISIi8iAGFhKJQ0LyZoaEePghERFJgIGFROZlzUoefkhERF6GgYVE9ky65db8REQkBQYWEtmzrJlzWIiISAqSB5Zt27YhLS0NcXFxmDRpEo4dO9Zs+T179mDEiBGIi4vDmDFjcPDgQYvn9+7di+nTp2PAgAGIjo7GyZMnLZ4vLy/HCy+8gN///veIj4/H7373O7z44ouorKy0KBcdHW31+PTTT13zob2UzmjuYWk5sNRylRAREXmQpIElJycH2dnZyMzMxK5duxATE4MZM2agpKTEZvkjR45g/vz5mDhxInbv3o2hQ4ciMzMTp0+fFstUV1cjOTkZTz75pM33uH79Oq5fv46FCxfik08+QXZ2Nr766issXrzYqmx2djYOHTokPoYNG+aaD+6l7NrpVsU5LERE5HlKKW++ceNGTJ48GRMmTAAAZGVl4cCBA9ixYwdmzZplVX7z5s0YNGgQZs6cCQCYN28ecnNzsXXrVixduhQAMHbsWADA5cuXbd6zd+/eeOONN8Tv77zzTsybNw9PPfUU9Ho9lMqGJmnXrh0iIyNb/TkNBmmHT8z3b6keuvoQooDQZFmlzDRspDUYodPpIW9mRdHtxt528nVsp5axjezjK+2kUCikrkKbIFlg0Wq1OHHiBGbPni1ek8vlSElJQX5+vs3XFBQUYNq0aRbXUlNTsW/fvlbVpaqqCsHBwRZhBTAFqMWLF6Nbt2744x//iAkTJkAmc/wf6MLCwlbVz1VaqkdldQ0A4Py5swiqsh34qhsNBX2bXwC/Znpjblfe8vvl7dhOLWMb2aett1P//v2lrkKbIFlgKSsrg8FgQHh4uMX18PBwnD171uZriouLERERYVW+uLjY6XqUlpbirbfewoMPPmhx/bHHHsO9996LgIAAHDp0CFlZWaiursbUqVMdvkdcXJykCdtgMKCwsLDFeii/OAigBn2if4XEO8NsltEZjMDuvQCAmL79EBqgckeVJWFvO/k6tlPL2Eb2YTuRIyQdEpJaVVUVZs+ejZ49e2Lu3LkWz2VmZopf9+3bFzU1NVi/fr1TgUWhUHjFD2NL9TCvEvJXqZosJ5fLIZcBRgHQG9tmV6e3/H55O7ZTy9hG9mE7kT0km3QbFhYGhUJhNcG2pKTEqhfFLCIiwqo3pbnyzamqqsLMmTMRFBSE1atXQ6VqvqcgISEBV69ehVardfhetwutuHFc08M8MplM3DyOK4WIiMhTJAssarUasbGxyMvLE68ZjUbk5eUhKSnJ5msSExNx+PBhi2u5ublITEx06N5VVVWYMWMGVCoV3n77bfj5+bX4mpMnTyI0NBRqtdqhe91O9MaWN44DGm/P37YnyhERkfeQdEgoIyMDCxcuRL9+/RAfH49NmzahpqYG48ePBwAsWLAAnTp1wvz58wEAU6dOxZQpU7BhwwYMGTIEOTk5OH78uLhCCDDts1JUVITr168DAM6dOwfA1DsTGRmJqqoqTJ8+HTU1NfjHP/6BqqoqVFVVAQA6dOgAhUKB/fv3o6SkBAkJCfDz88PXX3+NNWvWYPr06Z5sHo8zrxJqbh8WoPHmcexhISIiz5A0sIwaNQqlpaVYtWoVNBoN+vTpg3Xr1olDPEVFRZDLG/7xTE5OxvLly7FixQq8/vrr6N69O1avXo3evXuLZfbv34+nn35a/P6JJ54AAMydOxePPvooTpw4gaNHjwIA7r//fov6fPHFF4iKioJSqcS2bdvw8ssvAzAtfV60aBEmT57snobwEjpjy2cJAQ3nCbGHhYiIPEXySbfp6elIT0+3+dyWLVusro0cORIjR45s8v3Gjx8v9tDYMmDAAJw6darZOg0ePBiDBw9utkxbpLfjLCEA8DcPCXEOCxEReYjkW/OTdzAYBdR3sEAlt7eHhYGFiIg8g4GFADRsyw80v0oI4AGIRETkeQwsBMAysLQ46VbFAxCJiMizGFgIQMOmcYA9q4Q46ZaIiDyLgYUAALr6PVjkMkDRwoGGXNZMRESexsBCAACdwb4lzQDgr6rvYeGQEBEReQgDCwFotKS5hd4VgJNuiYjI8xhYCEDDpFuVsuU/EhwSIiIiT2NgIQCNhoRa2IMFAPxU5sMP2cNCRESewcBCABpWCalb2IMFYA8LERF5HgMLAQC09UNC9ky6FQMLJ90SEZGHMLAQgIZJty3tcgs0WiXESbdEROQhDCwEoGEOi9qRHhYOCRERkYcwsBCAho3j7Olh4eGHRETkaQwsBKBh0m1L2/IDDWcJcUiIiIg8hYGFADTah8WeZc1KHn5IRESexcBCABoCi2NDQuxhISIiz2BgIQANk24dGhJiDwsREXkIAwsBaHSWECfdEhGRF2JgIQCAzuhADwsPPyQiIg9jYCEAgE5v/063/iruw0JERJ7FwEIAAL3RvErI/iEhHn5IRESewsBCABycdNtop1tBENxaLyIiIoCBheo5tKy5/iwhQWgIOkRERO7EwEIAHNzpVtlQhhNviYjIExhYCECjnW7tWtbcOLBw4i0REbkfAwsBaBjasWeVkEwmg5onNhMRkQcxsBCAxj0s9v2RECfecqUQERF5AAMLAXBsWTPQeGkze1iIiMj9GFgIQKNlzUr7/kg0bB7HHhYiInI/BhYC0GhZs909LJzDQkREnsPAQgAcW9YM8ABEIiLyLAYWAgBoHZ10q+KkWyIi8hwGFgIA6B3Y6RbgkBAREXkWAwsBAPRG05CQ2s4eFn8VD0AkIiLPYWAhAIBWzx4WIiLyXgwsBKChh0Up56RbIiLyPgwsBKBhWbNa6WgPC4eEiIjI/RhYCECjs4Ts7WERVwmxh4WIiNyPgYUANKwS4j4sRETkjRhYCEDjww/tGxLi1vxERORJDCwEoNGQkIM9LDz8kIiIPIGBhQA43sPCSbdERORJDCwEoGFZs/1zWLgPCxEReQ4DCwFo3MNi7yqh+km3HBIiIiIPYGAhAA2BRSnnkBAREXkfBhYCAOgNjg0Jmc8S4pAQERF5AgMLQRCERnNYHOxh4eGHRETkAZIHlm3btiEtLQ1xcXGYNGkSjh071mz5PXv2YMSIEYiLi8OYMWNw8OBBi+f37t2L6dOnY8CAAYiOjsbJkyet3qOurg5ZWVkYMGAAkpKS8Oijj6K4uNiizJUrVzBr1iwkJCRg4MCBeOWVV6DX61v/gb2QeUkz4PiyZvawEBGRJ0gaWHJycpCdnY3MzEzs2rULMTExmDFjBkpKSmyWP3LkCObPn4+JEydi9+7dGDp0KDIzM3H69GmxTHV1NZKTk/Hkk082ed+XX34ZX375JVasWIEtW7bg+vXrmDt3rvi8wWDA7NmzodPp8MEHH2DZsmXYtWsXVq1a5boP70X0xobQobZ70i1XCRERkecopbz5xo0bMXnyZEyYMAEAkJWVhQMHDmDHjh2YNWuWVfnNmzdj0KBBmDlzJgBg3rx5yM3NxdatW7F06VIAwNixYwEAly9ftnnPyspK7NixA8uXL8fAgQMBmALMqFGjUFBQgMTERBw6dAhnzpzBxo0bERERgT59+uDxxx/H8uXLMXfuXKjVaoc+p8Eg7bCJ+f5N1aNW29BzJIMR9lTXfEZirc4g+edzlZbaiUzYTi1jG9nHV9pJoVBIXYU2wanAcunSJXTr1q1VN9ZqtThx4gRmz54tXpPL5UhJSUF+fr7N1xQUFGDatGkW11JTU7Fv3z6773v8+HHodDqkpKSI13r27IkuXbqIgaWgoAC9e/dGRESExX2WLFmCM2fOoG/fvnbfDwAKCwsdKu8uTdXjRm3DXxbHjx2FTNbyPJaLN3QAgJu1WhQUFLikft7CW36/vB3bqWVsI/u09Xbq37+/1FVoE5wKLPfffz9+/etfY+LEiRgxYgT8/Pwcfo+ysjIYDAaEh4dbXA8PD8fZs2dtvqa4uNgiRJjL3zr/pDnFxcVQqVRo166d1ftoNJom72P+3lzGEXFxcZImbIPBgMLCwibrUXSjFvj4AFQKGZKSkux6z7CSamDvf2GEDImJia6tsERaaicyYTu1jG1kH7YTOcKpwLJr1y7s2LEDy5YtwwsvvIBRo0Zh4sSJiI+Pd3X92gSFQuEVP4xN1UOAqUdFKZfbXc9APxUA0xwWb/hsruQtv1/eju3UMraRfdhOZA+nJt326dMHzz77LL766iu8/PLLuH79Ov785z/jgQcewMaNG1FaWtrie4SFhUGhUFhNsC0pKbHq3TCLiIiw6k1prnxT76HT6VBRUWH1PpGRkU3ex/y9uUxb4ug5QkDDsmadQYDBKLRQmoiIqHVatUpIqVRi+PDhWLVqFZ588klcuHABr7zyCoYMGYIFCxbg+vXrTb5WrVYjNjYWeXl54jWj0Yi8vLwmhyUSExNx+PBhi2u5ubkODUn069cPKpXK4r5nz57FlStXxPdJTEzE6dOnLcJUbm4ugoOD0atXL7vvdbvQObhpHNCwSggwTbwlIiJyp1atEiosLMSOHTuQk5ODgIAATJ8+HRMnTsS1a9fw5ptvYs6cOfjwww+bfH1GRgYWLlyIfv36IT4+Hps2bUJNTQ3Gjx8PAFiwYAE6deqE+fPnAwCmTp2KKVOmYMOGDRgyZAhycnJw/PhxcYUQAJSXl6OoqEgMS+fOnQNg6jWJjIxESEgIJkyYgGXLliE0NBTBwcF48cUXkZSUJAaW1NRU9OrVCwsWLMBTTz0FjUaDFStW4C9/+YvDK4RuB+K2/A70sPgrG7pva3UGBPlJuuCMiIjaOKf+ldm4cSN27tyJc+fOYfDgwWKvilxu+l93t27dsGzZMqSlpTX7PqNGjUJpaSlWrVoFjUaDPn36YN26deIQT1FRkfieAJCcnIzly5djxYoVeP3119G9e3esXr0avXv3Fsvs378fTz/9tPj9E088AQCYO3cuHn30UQDAM888A7lcjsceewxarRapqal4/vnnxdcoFAq88847WLJkCR588EEEBARg3LhxeOyxx5xpLq/n6MGHACCXy+CnlKNOb0QNe1iIiMjNnAos//rXvzBhwgSMGzcOHTt2tFmmQ4cOeOmll1p8r/T0dKSnp9t8bsuWLVbXRo4ciZEjRzb5fuPHjxd7aJri5+eH559/3iKk3Kpr165Yu3Zts+/TVjRsy+/YCGGAWoE6vZFDQkRE5HZOBZYNGzagS5cuFr0fgOlMmqKiInTp0gVqtRrjxo1zSSXJvRw9qdnMNCykQ62Ou90SEZF7OTXp9v7770dZWZnV9fLycgwdOrTVlSLPcmbSLWDqYQHAISEiInI7pwKLINhexlpdXe3UJnIkLb0Ty5oBwF9VH1i0DCxEROReDg0JZWdnAwBkMhlWrlyJgIAA8TmDwYBjx44hJibGtTUkt3Nm0i0ABNQvbWYPCxERuZtDgeWHH34AYOphOX36NFQqlficWq1GTEwMpk+f7toaktuZh4QcWdYMNPSwcNItERG5m0OBxbxq5+mnn8bixYsRHBzslkqRZ+mNzvawMLAQEZFnOLVKyDw0RG2DTu/cpFt/NeewEBGRZ9gdWObOnYtly5YhODgYc+fObbbsm2++2eqKkefojM4tazb3sNRwWTMREbmZ3YElJCTE5td0+9Pp64eElA72sNRPuuWQEBERuZvdgaXxMBCHhNoWcadbJ3tYGFiIiMjdnNqHpba2FjU1NeL3v/zyC9577z0cOnTIZRUjz2lYJeTcpFsuayYiIndzKrDMmTMHu3fvBgBUVFRg0qRJ2LhxI+bMmYP333/flfUjD3B2HxZOuiUiIk9xKrCcOHEC99xzDwDgP//5DyIiIvDll1/ilVdesXlgIXk3p3e6VdYPCek56ZaIqC3YvXs3BgwYAK1Wa3F9zpw5eOqppySqlYnTQ0JBQUEAgEOHDmH48OGQy+VITEzElStXXFpBcj9ta88SYg8LEVGbMGLECBgMBnzxxRfitZKSEhw8eBATJkyQsGZOBpY777wT+/btQ1FREQ4dOoTf/va3AEwfipvJ3X7MPSyO7nTLSbdERG2Lv78/HnjgAezcuVO89tFHH6Fz584YMGCAhDVzMrBkZmbi1VdfRVpaGhISEpCUlAQA+Prrr9GnTx+XVpDcr2GVkKPLmjnploiorZk8eTK+/vprXLt2DQCwc+dOjBs3DjKZY/+pdTWndrodMWIE+vfvD41GY3HY4cCBAzFs2DCXVY48Q+vspFvuw0JE1Ob07dsXMTEx2L17N37729/izJkzGD9+vNTVci6wAEBkZCQiIyMtrsXHx7e6QuR5rR0SYg8LEVHbMnHiRGzatAnXrl1DSkoKOnfuLHWVnAss1dXVePfdd3H48GGUlJTAaLRcJdJ4sg55P/M+LGonJ93WctItEVGbMmbMGLz66qvYvn07Xn31VamrA8DJwPLss8/im2++wf/93/8hMjJS8nEtah0de1iIiKiRkJAQDB8+HAcPHvSaqR5OBZb//ve/WLNmDfr37+/q+pAE9E7udOsvrhLiPixERG3NtWvXMGbMGKjVaqmrAsDJVULt2rVD+/btXVwVkoq5h0Xt6MZxjXpYBEFweb2IiMjzbty4gc8//xzffPMN/vznP0tdHZFTgeXxxx/HypUrLc4TotuXrn5Zs9LBZc3mOSwAUMfdbomI2oRx48Zh0aJFePLJJ9GjRw+pqyNyakho48aNuHjxIlJSUhAVFQWl0vJtdu3a5ZLKkWfo6sOGSungkFCj8jVag9jjQkREt6/9+/dLXQWbnAos3jIBh1xDX7/KSyV3bEhIqZBDpZBBZxBQq+fEWyIich+nAsvcuXNdXQ+SkM7JSbeAaR6LzqDneUJERORWTs1hAYCKigr8+9//xmuvvYby8nIAplOczVv50u1D5+RpzQCXNhMRkWc41cPy448/IiMjAyEhIfjll18wefJktG/fHnv37kVRUZHXbDJD9tE7eVoz0GjzOAYWIiJyI6d6WJYtW4Zx48Zh7969FuuzhwwZgu+++85llSPP0Dl5lhAA+Cu5FwsREbmfU4GlsLAQf/zjH62ud+rUCRqNptWVIs/SGZ3b6RYA/Ot7WDiHhYiI3MmpwKJWq1FVVWV1/fz58+jQoUOrK0WeJQ4JObgPCwAE1J/YzDksRETkTk4FlrS0NKxevRo6nU68duXKFSxfvhzDhw93WeXIM8QhISUn3RIRkWtt27YNaWlpiIuLw6RJk3Ds2DGn3sepwLJo0SJUV1dj4MCBqKurw5QpUzB8+HAEBQXhiSeecKoiJB1xWbMTPSzmzeLqGFiIiOgWOTk5yM7ORmZmJnbt2oWYmBjMmDEDJSUlDr+XU6uEQkJCsHHjRnz//ff48ccfUV1djdjYWKSkpDjzdiSxhrOEnBkSYg8LEZEUBEHw6N+9ASoFZDLHeuI3btyIyZMnY8KECQCArKwsHDhwADt27MCsWbMcei+HA4vRaMTOnTvx+eef45dffoFMJkPXrl0RGRkJQRAc/jAkvYbTmlsz6ZarhIiIPEUQBEx8Jw/fXyjz2D3vuSsM/354oN3/zmu1Wpw4cQKzZ88Wr8nlcqSkpCA/P9/h+zv0X2pBEPDII4/g2WefxbVr19C7d2/06tULV65cwaJFi5CZmelwBUhagiC0apUQe1iIiKTh7d0DZWVlMBgMCA8Pt7geHh6O4uJih9/PoR6WnTt34ttvv8V7772He++91+K5vLw8ZGZmYvfu3Rg7dqzDFSFpGIwCBFMHi1NDQv71q4S4cRwRkefIZDL8++GBXj8k5EoOBZZPP/0UDz/8sFVYAYCBAwdi1qxZ+PjjjxlYbiN6oyB+7cxZQuYeFgYWIiLPkslkCFQ7NRXVI8LCwqBQKKwm2JaUlCAiIsLh93PoX6hTp05h0KBBTT4/ePBg/Pjjjw5XgqSjNTTMPXHmLCF/DgkREZENarUasbGxyMvLE68ZjUbk5eUhKSnJ4fdzKJrduHHDaiyqsfDwcNy4ccPhSpB0zBNuASc3juNOt0RE1ISMjAwsXLgQ/fr1Q3x8PDZt2oSamhqMHz/e4fdyKLAYDAYolU2/RKFQwGDgP1y3E319D4tcBsjlTvSwmM8S0nOVEBERWRo1ahRKS0uxatUqaDQa9OnTB+vWrXNqSMihwCIIAhYtWmRx4GFjWq3W4QqQtLStOPgQaHRaM3tYiIjIhvT0dKSnp7f6fRwKLOPGjWuxDCfc3l7Ec4ScDSycw0JERB7gUGDJzs52Vz1IIuI5Qk5MuAU46ZaIiDzDuf9WU5shniPkZA8L92EhIiJPYGDxcfr6XW5VTky4BRrNYWFgISIiN2Jg8XHikJCylXNYOOmWiIjciIHFx4lDQs72sDSawyIIQguliYiInOMVgWXbtm1IS0tDXFwcJk2ahGPHjjVbfs+ePRgxYgTi4uIwZswYHDx40OJ5QRCwcuVKpKamIj4+HtOmTcP58+fF5//3v/8hOjra5sN878uXL9t8vqCgwNUfX1KtXSXkVx9YjEJD+CEiInI1yQNLTk4OsrOzkZmZiV27diEmJgYzZsywOnvA7MiRI5g/fz4mTpyI3bt3Y+jQocjMzMTp06fFMmvXrsWWLVuwZMkSbN++HQEBAZgxYwbq6uoAAElJSTh06JDFY9KkSYiKikJcXJzF/d577z2LcrGxse5rDAnoWrsPS31gAbhSiIiI3EfyU5M2btyIyZMnY8KECQCArKwsHDhwADt27MCsWbOsym/evBmDBg3CzJkzAQDz5s1Dbm4utm7diqVLl0IQBGzevBmPPPIIhg0bBgB49dVXkZKSgn379mH06NFQq9WIjIwU31On0+GLL75Aenq61UmU7du3tyjrDKl3/zXf31Y96nR6AIBS7lw95RCgkMtgMAq4WatFsFryDOy05tqJGrCdWsY2so+vtJNCoWi5ELVI0sCi1Wpx4sQJzJ49W7wml8uRkpKC/Px8m68pKCjAtGnTLK6lpqZi3759AExDORqNBikpKeLzISEhSEhIQH5+PkaPHm31nvv370d5ebkYmhp75JFHUFdXh+7du2PmzJkYOnSow5+zsLDQ4de4g616/HSpFgBQV1Pt9HCXWg7UGIHvjx5HlxDJM3Crecvvl7djO7WMbWSftt5O/fv3l7oKbYKk/7qUlZXBYDBYHagYHh6Os2fP2nxNcXGx1RkE4eHhKC4uBgBoNBrxWlNlbvXhhx8iNTUVd9xxh3gtMDAQixYtQnJyMmQyGfbu3YvMzEysXr3a4dASFxcnacI2GAwoLCy0WY+LsivA4XK0D22HxMREp94/5LMvUVNZh+49e6Nvl3YuqLE0mmsnasB2ahnbyD5sp7bv22+/xfr163H8+HFoNBqsXr1aHP1w1O3/3+FWunr1Kg4dOoQVK1ZYXO/QoQMyMjLE7+Pj43H9+nWsX7/e4cCiUCi84ofRVj0MgmkITK2UO13HQPNeLAbBKz5na3nL75e3Yzu1jG1kH7ZT21VdXY3o6GhMmDABc+fObdV7SRpYwsLCoFAorCbYlpSUNHmSY0REhFVPSePy5vkmJSUl6Nixo0WZmJgYq/fbsWMH2rdvj7S0tBbrm5CQgNzc3BbL3U7Mk26VcufnngSqTX+MqrkXCxGR5wgCoKv23P1UgYDMsS0whgwZgiFDhrjk9pIGFrVajdjYWOTl5YldREajEXl5eU2e7JiYmIjDhw9bzGPJzc0VhzOioqIQGRmJvLw89OnTBwBQVVWFo0eP4k9/+pPFewmCgJ07d2Ls2LFQqVQt1vfkyZOtnoDrbfStPEsIaOhhqa7Tu6RORETUAkEANvweuPQ/z92z273A9M8cDi2uIvmQUEZGBhYuXIh+/fohPj4emzZtQk1NDcaPHw8AWLBgATp16oT58+cDAKZOnYopU6Zgw4YNGDJkCHJycnD8+HEsXboUACCTyTB16lS8/fbbuOuuuxAVFYWVK1eiY8eOVuNmhw8fxuXLlzFx4kSreu3atQsqlUoMPZ9//jl27NiBF1980Z3N4XHaVu7DAgCBfuxhISLyPGmCg1QkDyyjRo1CaWkpVq1aBY1Ggz59+mDdunXiEE9RURHkjYYrkpOTsXz5cqxYsQKvv/46unfvjtWrV6N3795imYceegg1NTV47rnnUFFRgf79+2PdunXw8/OzuPeHH36IpKQk9OzZ02bd3nrrLVy5cgUKhQI9evTAP//5T4wYMcINrSAdcw+LsjU9LPV7sVRr2cNCROQRMpmpt8PLh4RcSfLAAgDp6elNDgFt2bLF6trIkSMxcuTIJt9PJpPh8ccfx+OPP97sfV977bUmnxs3bhzGjRvX7OvbAr3R1MOiblUPizmwsIeFiMhjZDJAHSR1LTzm9t3li1xCq299D0tQ/aTbmwwsRETkJl7Rw0LS0RtdsUrIfGIzh4SIiKjBzZs3cfHiRfH7y5cv4+TJkwgNDUWXLl0cei8GFh9nPrBQrWz9smb2sBARUWPHjx/H1KlTxe+zs7MBmKZdLFu2zKH3YmDxcQ37sLR+WXMNAwsRETUyYMAAnDp1yiXvxTksPk7vkmXNpsByk/uwEBGRmzCw+DidCzeOq9Gxh4WIiNyDgcXHmeewKFvTw2Kew8IeFiIichMGFh/X0MPi/B+FIJ4lREREbsbA4uPMy5pbMyQUoObGcURE5F4MLD5O54JJt0Hc6ZaIiNyMgcXHuWRZs8o8JMQ5LERE5B4MLD7Olcuaq7UGGOvPJiIiInIlBhYfp3XBpFvzsmYAqNVzWIiIiFyPgcXH6Q2tP/zQX6kQTxy/WcfAQkRErsfA4uP09UM46lb0sMjlMgSouD0/ERG5DwOLj9PqW9/DAjRsHlfFzeOIiMgNGFh8nLmHRSlv3R+FYPN5QlwpREREbsDA4uPMy5rVytb1sAT7s4eFiIjch4HFx5mXNbe2hyWI5wkREZEbMbD4OFecJQQAwX4MLERE5D4MLD6uIbC4ZkiospaBhYiIXI+BxceJQ0Kt7GEJEntYuKyZiIhcj4HFx2ld1cNiDixcJURERG7AwOLjzMuaXTWHhUNCRETkDgwsPsxoFGBwUWAJ4qRbIiJyIwYWH6YzGsWvW7vTrbhxHAMLERG5AQOLDzNPuAUAVWv3YTEPCTGwEBGRGzCw+DDzkmbAhZNuGViIiMgNGFh8mK5RD4tCzsBCRETei4HFh+nr57CoFXLIZK0LLOYhoSruw0JERG7AwOLDdHrzpnGtCytAQw9LVZ2u1e9FRER0KwYWH2ZeJaRs5XAQ0BBYanVG6BvNjSEiInIFBhYfZp50q1a2/o+BeUgIAG5qOSxERESuxcDiw8RzhFq5pBkwhR51/eZzVZx4S0RELsbA4sPEk5qVrR8SAhpObOZKISIicjUGFh9mXtbc2k3jzILqd7vleUJERORqDCw+zDw51hWrhAAgxE8FAKis5UohIiJyLQYWH6Y1Dwm18uBDs3YBPLGZiIjcg4HFh4mTbl0UWEL8TT0sFexhISIiF2Ng8WENO926aEjInz0sRETkHgwsPkzrwmXNANDOn3NYiIjIPRhYfJirJ922Yw8LERG5CQOLDxN3unX1HJYa9rAQEZFrMbD4MJ3BdYcfApzDQkRE7sPA4sP0Ll7WHCLOYWFgISIi12Jg8WHiTrcuCyymHhYuayYiIldjYPFhuvplzUq5iybdBrCHhYiI3MMrAsu2bduQlpaGuLg4TJo0CceOHWu2/J49ezBixAjExcVhzJgxOHjwoMXzgiBg5cqVSE1NRXx8PKZNm4bz589blElLS0N0dLTF491337Uo8+OPP+LPf/4z4uLiMGTIEKxdu9Yln9db6PT1PSxK9rAQEZF3kzyw5OTkIDs7G5mZmdi1axdiYmIwY8YMlJSU2Cx/5MgRzJ8/HxMnTsTu3bsxdOhQZGZm4vTp02KZtWvXYsuWLViyZAm2b9+OgIAAzJgxA3V1dRbv9dhjj+HQoUPiIz09XXyuqqoKM2bMQJcuXbBz504sWLAAb775Jv7f//t/7mkICZg3jlO5qIfFHFiq6vQwGgWXvCcRERHgBYFl48aNmDx5MiZMmIBevXohKysL/v7+2LFjh83ymzdvxqBBgzBz5kz07NkT8+bNQ9++fbF161YApt6VzZs345FHHsGwYcMQExODV199FdevX8e+ffss3isoKAiRkZHiIzAwUHzuo48+gk6nw8svv4xf/epXGD16NKZMmYKNGze6rzE8zNVzWMwbxwkCUKXlsBAREbmOUsqba7VanDhxArNnzxavyeVypKSkID8/3+ZrCgoKMG3aNItrqampYhi5fPkyNBoNUlJSxOdDQkKQkJCA/Px8jB49Wry+du1avP322+jcuTMeeOABTJs2DUqlUrzPPffcA7VabXGftWvX4saNGwgNDbX7cxoMBrvLuoP5/rfWQ6s3hQqF3DV1VMoEqBUyaA0CbtysQ5BK8jzskKbaiSyxnVrGNrKPr7STQqGQugptgqSBpaysDAaDAeHh4RbXw8PDcfbsWZuvKS4uRkREhFX54uJiAIBGoxGvNVUGAKZMmYK+ffsiNDQU+fn5eP3116HRaPD000+L94mKirJ4D/N9i4uLHQoshYWFdpd1p1vrUXStAgBQormOgoIal9zDX2kKLN8UHEf39iqXvKenecvvl7djO7WMbWSftt5O/fv3l7oKbYKkgUVKGRkZ4tcxMTFQqVR4/vnnMX/+fIteFVeIi4uTNGEbDAYUFhZa1SP0/HEA1Yjq0hmJib1ccq+w/f9FRV01unbvicTuHVzynp7SVDuRJbZTy9hG9mE7kSMkDSxhYWFQKBRWE2xLSkqselHMIiIiLHpKbi0fGRkpXuvYsaNFmZiYmCbrkpCQAL1ej8uXL6NHjx4272P+vqm6NUWhUHjFD+Ot9ajfNw5qlevqF1q/tLmqzugVn9kZ3vL75e3YTi1jG9mH7UT2kHSSgVqtRmxsLPLy8sRrRqMReXl5SEpKsvmaxMREHD582OJabm4uEhMTAQBRUVGIjIy0eM+qqiocPXq0yfcEgJMnT0Iul4tDSYmJifjuu++g0zUs0c3NzcXdd9/t0HCQN9PXr+Rx1VlCQENgKed5QkRE5EKSz4rMyMjA9u3bsWvXLvz8889YsmQJampqMH78eADAggUL8Nprr4nlp06diq+++gobNmzAzz//jDfeeAPHjx8XlyTLZDJMnToVb7/9Nr744gucOnUKCxYsQMeOHTFs2DAAQH5+Pt577z38+OOPuHTpEj766CNkZ2fjD3/4gxhGxowZA5VKhcWLF+Onn35CTk4ONm/ebDGUdLvTGly7cRzQEFhuMLAQEZELST6HZdSoUSgtLcWqVaug0WjQp08frFu3Thx2KSoqglzekKuSk5OxfPlyrFixAq+//jq6d++O1atXo3fv3mKZhx56CDU1NXjuuedQUVGB/v37Y926dfDz8wNg6tnJycnBm2++Ca1Wi6ioKEybNs0ijISEhGD9+vVYunQpxo8fj7CwMMyZMwcPPvigh1rG/cxnCSnd0MPCwEJERK4keWABgPT0dItN2xrbsmWL1bWRI0di5MiRTb6fTCbD448/jscff9zm87Gxsdi+fXuL9YqJicH777/fYrnblXkfFncMCVUwsBARkQtJPiRE0tGJPSwcEiIiIu/GwOLDtHpTYFG76CwhgIGFiIjcg4HFh5l7WFy1NT/AwEJERO7BwOLDxDksruxhCWRgISIi12Ng8WHikBB7WIiIyMsxsPgwtw4JVesgCILL3peIiHwbA4sPM28c545Jt1qDEbU6o8vel4iIfBsDiw8zDwmpXLisOdhPCUX9zrkcFiIiIldhYPFh5iEhV85hkclkaOdv2o+QgYWIiFyFgcWHuWMfFgBoH6gGwMBCRESuw8Diw8zLml056RYA2plPbK7WuvR9iYjIdzGw+ChBENwy6RYAwgLNgYU9LERE5BoMLD7K3LsCuL6HpUP9kFAZe1iIiMhFGFh8lHnCLeDaSbdAwxyWUgYWIiJyEQYWH2URWFw8JNQhqH5I6CaHhIiIyDUYWHyUeYWQXAZx3xRXYQ8LERG5GgOLj9K6YVt+sw5BpsDCVUJEROQqDCw+yh0nNZu1r18lVMZVQkRE5CIMLD7KHSc1m5l7WMpusoeFiIhcg4HFR7njpGazsPo5LOU1OhiNPLGZiIhaj4HFR7lr0zigYUjIYBRQWat3+fsTEZHvYWDxUe44qdnMT6lAkFoBgJvHERGRazCw+CjxpGalwi3vHxbEpc1EROQ6DCw+qmHSret7WIBG81gYWIiIyAUYWHyUOyfdAg0rhYqrGFiIiKj1GFh8lNaN+7AAQESwHwCguKrOLe9PRES+hYHFRzVMunVTYAmp72GpZA8LERG1HgOLj3L3kFAke1iIiMiFGFh8lDmw+HFIiIiIbgMMLD7KnfuwAAwsRETkWgwsPsqdpzUDDXNYNJUMLERE1HoMLD5Kp/fMKqGyap04/EREROQsBhYfpTUYALivhyUsUA2F3DTcVMpTm4mIqJUYWHyUzs37sCjkMnHzOA4LERFRazGw+KiGrfnd90eAE2+JiMhVGFh8lLsn3QJARDB7WIiIyDUYWHyUztzD4qYhIQC4o50/AOBaRa3b7kFERL6BgcVHNfSwuGcfFgDoHGoKLFduMLAQEVHrMLD4KPNSY3f2sHRuHwAAuMrAQkRErcTA4qO05n1Y3DiH5Y76HpYiBhYiImolBhYf5YlJt53FwFLjtnsQEZFvYGDxUeZJtyp3DgmFmoaEyqt1qNEa3HYfIiJq+xhYfJS5h8WdQ0Lt/JUIVCsAsJeFiIhah4HFR9XqTD0e/ir3/RGQyWTisBAn3hIRUWswsPiouvohIT+lwq33MQ8LcWkzERG1BgOLj/JEDwsAdGlv6mG5XFbt1vsQEVHbxsDio2p1ph4Wf5V7e1ju7BAIALhUyjksRETkPAYWH1WnN/Ww+LlxlRAAdBMDC3tYiIjIeV4RWLZt24a0tDTExcVh0qRJOHbsWLPl9+zZgxEjRiAuLg5jxozBwYMHLZ4XBAErV65Eamoq4uPjMW3aNJw/f158/vLly3jmmWeQlpaG+Ph4DBs2DKtWrYJWq7UoEx0dbfUoKChw5UeXTJ2HeljuCg8CAFxkYCEiolaQPLDk5OQgOzsbmZmZ2LVrF2JiYjBjxgyUlJTYLH/kyBHMnz8fEydOxO7duzF06FBkZmbi9OnTYpm1a9diy5YtWLJkCbZv346AgADMmDEDdXWmU4PPnj0LQRCwdOlSfPrpp3j66afxwQcf4J///KfV/d577z0cOnRIfMTGxrqnITzIYBTEZc3u7mExDwldragV580QERE5SvLAsnHjRkyePBkTJkxAr169kJWVBX9/f+zYscNm+c2bN2PQoEGYOXMmevbsiXnz5qFv377YunUrAFPvyubNm/HII49g2LBhiImJwauvvorr169j3759AIDBgwcjOzsbqamp6NatG4YOHYrp06dj7969Vvdr3749IiMjxYdKpXJfY3iItn6FEOD+HpawQBWC/ZQAgMtlnMdCRETOUUp5c61WixMnTmD27NniNblcjpSUFOTn59t8TUFBAaZNm2ZxLTU1VQwjly9fhkajQUpKivh8SEgIEhISkJ+fj9GjR9t838rKSoSGhlpdf+SRR1BXV4fu3btj5syZGDp0qKMfEwaDtD0L5vubf71Z1zD0pZQJbq9ft7AAnLxaifPFVbg7PMCt92qNW9uJbGM7tYxtZB9faSeFwr3/MfQVkgaWsrIyGAwGhIeHW1wPDw/H2bNnbb6muLgYERERVuWLi4sBABqNRrzWVJlbXbhwAVu3bsXChQvFa4GBgVi0aBGSk5Mhk8mwd+9eZGZmYvXq1Q6HlsLCQofKu4u5HiXVpr8cFDLgeGHz84VcoZ3CFJDyjp1GWM0vbr9fa3nL75e3Yzu1jG1kn7beTv3795e6Cm2CpIHFG1y7dg0zZ87EiBEjMHnyZPF6hw4dkJGRIX4fHx+P69evY/369Q4Hlri4OEkTtsFgQGFhoViP8yU3gU81CFArkJiY6Pb7J1w9hf/9cg66gDAkJvZ1+/2cdWs7kW1sp5axjezDdiJHSBpYwsLCoFAorCbYlpSUWPWimEVERFj1lDQuHxkZKV7r2LGjRZmYmBiL1127dg1Tp05FUlISXnjhhRbrm5CQgNzc3JY/2C0UCoVX/DCa61G/QAj+Ks/Uq1fHEADAueJqr2iHlnjL75e3Yzu1jG1kH7YT2UPSSbdqtRqxsbHIy8sTrxmNRuTl5SEpKcnmaxITE3H48GGLa7m5uWJPQVRUFCIjIy3es6qqCkePHrV4T3NYiY2NRXZ2NuTylpvi5MmTYiC6nZmXNLt7W36zHpGmpc1nNTc9cj8iImp7JB8SysjIwMKFC9GvXz/Ex8dj06ZNqKmpwfjx4wEACxYsQKdOnTB//nwAwNSpUzFlyhRs2LABQ4YMQU5ODo4fP46lS5cCMB24N3XqVLz99tu46667EBUVhZUrV6Jjx44YNmwYAFNYmTJlCrp06YKFCxeitLRUrI85kOzatQsqlQp9+vQBAHz++efYsWMHXnzxRY+1jbuYlxf7uXlbfrOekcEAgF/Ka1CjNSBAzf9JERGRYyQPLKNGjUJpaSlWrVoFjUaDPn36YN26deIQT1FRkUXvR3JyMpYvX44VK1bg9ddfR/fu3bF69Wr07t1bLPPQQw+hpqYGzz33HCoqKtC/f3+sW7cOfn5+AICvv/4aFy5cwIULFzB48GCL+pw6dUr8+q233sKVK1egUCjQo0cP/POf/8SIESPc2RweUeuhgw/NwoLUCAtUoaxah7PFVYjtYr0ai4iIqDmSBxYASE9PR3p6us3ntmzZYnVt5MiRGDlyZJPvJ5PJ8Pjjj+Pxxx+3+fz48ePFHpymjBs3DuPGjWu2zO2qzkMHHzbWMzIY310ow1nNTQYWIiJymOQbx5HnmXtY/D3UwwI0zGP5WVPlsXsSEVHbwcDigzw9hwUAetTPY+HEWyIicgYDiw+qk6CHxTzxlj0sRETkDAYWHyTFHJbGS5uNRsFj9yUioraBgcUHiUNCHuxhubNDIJRyGWp0BlytqPXYfYmIqG1gYPFB4pCQB3tYVAo57gwPBMB5LERE5DgGFh/UMOnWsxu49YjgPBYiInIOA4sPaph069nf/p4dubSZiIicw8Dig6TqYelVv1LozHUGFiIicgwDiw+qFQ8/9Oxv/686mU5t/omBhYiIHMTA4oPq9OZlzR7uYelo6mHRVNahvFrr0XsTEdHtjYHFB5l7WDwdWIL9lOgS6g+Aw0JEROQYBhYf1LAPi+d/+3txWIiIiJzAwOKDGvZh8WwPCwD8qn5Y6KdrDCxERGQ/BhYfVKM19bAESBBYout7WH68WuHxexMR0e2LgcUHVdXpAQDB/kqP37tvl3YAgB+KKiAIPFOIiIjsw8Dig8TA4ifBkFCnYCjlMpRX61B0g2cKERGRfRhYfIwgCI0Ci8rj9/dTKsTlzSeucFiIiIjsw8DiY+r0RhiMpqEYKYaEgEbDQgwsRERkJwYWH1NZqxe/DpRg0i0AxHYJBQAcvVwuyf2JiOj2w8DiYxqGg5SQy2WS1GHA3R0AAN+cK4XeYJSkDkREdHthYPExNxsFFqn07dwO7QNVqKrT4+jlG5LVg4iIbh8MLD7GPCQUJMEKITO5XIaBPcIBALlniiWrBxER3T4YWHxMwx4snl8h1Nhve0UAAHKOX+V+LERE1CIGFh9jHhIKkXBICAAeiO8Mf5UcJ4sq8M25UknrQkRE3o+BxcdU1kk/JAQA7QPVGJcUBQB4++DPktaFiIi8HwOLj6mqlW7TuFs9NOhuqBQyHDilwec/XJO6OkRE5MUYWHyMOCQk0aZxjfWIDMbMQT0AAC99+gOXOBMRUZMYWHxMlZcMCZnNva8XOgSpcb6kGrvyf5G6OkRE5KUYWHxMpRcNCQFAkJ8SswebelneOfgzVwwREZFNDCw+Rtw4zguGhMz+cu9dCFAp8LPmJvIvlUtdHSIi8kIMLD6mYWt+7xgSAky77o7sdwcAYMf3lyWuDREReSMGFh9TWeddQ0JmE/qbljh/fPQKanUGiWtDRETehoHFx1TV6gB4z6Rbs4E9wtEl1B8VtXrsO8klzkREZImBxccUV2kBABHBfhLXxJJcLsP4ZFMvC4eFiIjoVgwsPqROZ8CNGlMPS8cQ7wosADA+uSsA4L8/FeN6Za3EtSEiIm/CwOJDNFV1AAC1Uo7QAO+awwKYNpJLvrM9DEYB/1/+FamrQ0REXoSBxYdcrzQFlo4hfpDJZBLXxraJ/bsBALZ/dwlGI/dkISIiE+/ZjIPc7npFQ2DxVqPjO+PlnJP46XoVPtr/X4ztcAEoPg2UngNqyoG6CtPDoAPkSkChAhRqwK8dENgBCAirf7Rv+No/FPALBfzb1X/dDlAFAF4a2oiIyBoDiw9p6GHxl7gmNhh0wMU8hJ49gEeCqvGPukHI3ncR9/o9hztkZe65p1xZ/1ABCiXkciXi9XrIv1QDkNUHGlmjYNP42i3fA6bwpA4G1EGmX/0af90oUAWGm74ODAcCOgDqQPd8PiKiNoSBxYeIgaWdl/SwCAJw+Tvg2P8DTuwEqksAADMEFXbIeuKs0AV/Fl7CG/1+RuxdnU3/yPuFAn4hpnBgNABGHaCvM/W61JQB1aWmXxs/aiuAuhv1v1YAQv0hi0a96QHTBF8ZABUA1Hm4HZQBps8WFAG06wqEdAbadW70dRfTwy/EwxUjIvIeDCw+RFMfWDq1k7iHpa4KOPov4H/vACVnGq4HRgC9hsK/x++wKexeTNp2Dmcr2mNswT2YF9Ebs/v3gFLRymlXggBoqwBdrSnsGHSm0GLQwaDX4tSPPyI6ujcUcpmpLISG10Go/1Zo9H2jMgYtoL1Z/6gyfU7z17U3gJpSoLrMFMxqSk3hyqgD9DVAxS+mR9HRpuuuDgFCuwKhUfWPbvWP+u/bdTEFOSKiNoiBxYeYe1gipZrDUn4J+OZd4Mgm0z/gAKAKBPqMAeImAz1+ByhMfyS7Afjksa5YvKsQ/zlxDf/4zyl8+eN1rJ/269atcJLJTD0VtnorDAbUFOmAO+IAhQc21hMEoK6yPryUAFUaU2ipLAIqiiy/rrsBaCsBzY+mhy0yualHRgw0UY0CTf2v/qGcu0NEtyUGFh/SeJWQR136Fji8GvjhI0Co33a/Qw9gwCNA4p+aHOqICPbDO+n9sfPIL1jy0Ql8d6EMD2/5Hpum/wZqZRtY4CaT1U8EbgeEdW++bF0VUHHFFGJuXAZuXLrl18umHh5zT82l/9l+H3MvTVBk/SPC1LMVFF7/awTg374h1Pm1E0MkEZGU+DeRjzAIAi6XVQMAurQPcP8N9Vrg5EemYZ/L3zZcv3swcO8c4Fe/B+Qthw6ZTIYJ/aMQ0zkEk9/JQ97ZEvzjPz9i8ei+bqy8F/ILBiJ7mx62GI3ATY11iLlxqeH76pKWe2lsUQU2BBhVAOQKNXrX6CA/EQ4o/QGln+kha9Qr1XiiMmCaNyQYTPOObv3aaGj0vcHU83TrNYuy9a+XKUyrvVSB9b82+tovxBS8zKvFbv3aP5RBjOg2w59YH/FLhR5VdQYEqhXoGRnsvhtVXQe+fw/4dj1QddV0TaEG4iYB9z5iGm5xQmyXULz+YCJmb/kea786h9/cHY77+3ZyXb1vd3I5ENLJ9Ijqb7uMtrq+h+YScLPY9Kg2/1rS8H1thWmoSl9jep2u2vSoMp3xJAMQAgClnvhgbuQfCgR1bOhpCm70tXg9EgiONPU0cSiNSFIMLD7idKlpS/64rqGmCaWuZDQCF/OA/C3A8R2moQkACO4E3DMDuCfD9I9BK/0+9g5M/+3d2PD1OczfXoAdj6TgV524csZu6kAg4lemhz302oYJw3WVYogxaGtx4edT6B7VGXKj1lROX9uw+gqAOBEZMPWYyOSAXGHqFRG/bnTN/L1MYQpf4rVmyhv1psnTumpAV9Po1xrTarDactPePTVljb4uN/UyAabPVXsDKPmp5bZQBVqu2GrXBQjpUr+aq/7r4I6mehGRWzCw+IifSkyBJenOMNe8oSAA138wBZRj/wZuXGx4rmt/0/yUvv8HKNWuuV+9RSNjcORiGQoulWPiO3l48NfdoFbIcamsGldv1CLYT4lenYIR1zUU8V3bo1uHAK/d1dfrKdWAsoNpyXVjBgPKajrjrvhEz0xOdjWDrj68lJqG0aqu1/c4aWw8ik3hR1cNlP5sejRFpgBC7gDadYE8+A5E1akgq0loWMFlDjxKL9lWgOg24xWBZdu2bVi/fj00Gg1iYmLw97//HfHx8U2W37NnD1auXIlffvkF3bt3x5NPPokhQ4aIzwuCgFWrVuHf//43KioqkJycjCVLlqB79+5imfLycrzwwgv48ssvIZfLMXz4cCxevBhBQUFimR9//BFLly5FYWEhOnTogPT0dDz00ENuaQN3O1UfWBK7tXf+Tao0wKXDwE+fA2f2mYYXzNQhpoByTwYQdU/rKtsMtVKOjdN+jYz3vkXBpXK8+9+zVmW++PG6+HVogAr9urbDr7t3wJDekYjrGtr6pdF0e1OoTMM8wZFAZHTL5XU19au1rtyyeuuX+u+vmIY/BYM46VkGoBMAnNtp/X6BEbfss9PVFHTME6DNe/JwGIrIgkwQBEkPbMnJycGCBQuQlZWFhIQEbNq0CZ999hk+++wzhIeHW5U/cuQI0tPT8be//Q333XcfPv74Y6xbtw47d+5E796mCYnvvvsu3n33XSxbtgxRUVFYuXIlTp8+jZycHPj5mf53M3PmTGg0GixduhQ6nQ7PPPMM4uLi8NprrwEAqqqq8Pvf/x4DBw7E7Nmzcfr0aTzzzDN45pln8OCDD9r12QwGAwoKCpCYmAiFhP8T3ffDVczc/D1UChnynh6KiOAW/odXWwGUnQNKzwIlPwNXjwG/5Fv2ogCmCZc97gPiJwPRI02THT2kTm/AnsKrOHy2BCqFHF3DAtA51B+VtXr8UFSB47/cwI9FldAajBavUyvkuDsiCJEhflAr5VApZFArFVDJZRAgoKy0FOHh4ZDJZNDqjdDqjdAZjNAajKir/94oCIgKC0DMHe0QfUcI2geoIJebyldrDajRGVCj1UOrN0JvFOCnVCAqLADdw4MQFRYAuauH5DzMW/5cexWDHrh5vT7UXIHxxmVcP3MUHQP0kIth5wpgcGBXQrmqfldk887IHUwhxj+04YgJ/3Y2fg017bCs9PP6wMM/S+QIyQPLpEmTEBcXh+eeew4AYDQaMWTIEEyZMgWzZs2yKj9v3jzU1NRgzZo14rXJkycjJiYGS5cuhSAIGDRoEDIyMjBjxgwAQGVlJVJSUrBs2TKMHj0aP//8M0aNGoUPP/wQcXGmSaD//e9/MWvWLBw8eBCdOnXC+++/jxUrVuDQoUNQq03DGsuXL8e+ffvw2Wef2fXZXPLDWHnN1KshGC03KxOE+jkDTX/90w05vriixoYzgbheq8BDPcuxOLrINN9AV2P6ta6yYdJldalp0qWuuonKyEz/I+3xO6DX/UD333o0pDhKqzfi9LVKHL1cjkM/FePQT8WorNNLWqcAlQK9OgajR2QQ2geoEOKvQoi/EsH+SggCoDOYQk9lrR5VdToYBUAuA+QyGeQyGW7U6HD1Ri2uVZh2570j1B9d2gcgKiwAHUP84Ve/3LuyVofyGh2Kq+pQXKnFjRodAtQKhPgr0a7+niH+KhgFAXU6A+r0RtTqDKjVGVGnN/1aqzdAEEx1DlAr4K+Sw1+lgL9SjvLi6+gW1QW1OiOq6gyo0enhp1QgUG1+KBHkp0CAWgkZTJ/LFPwE6PRG6I1G6AyCGAJ1BqMYEA2CgABVw/sEqhUI9FMgQKWAVm9Ejc6Aaq0BtfW/1uhMX9fpjeJnqav/HEYBYp0C1Aoo5XIoFTKozL8qTKHVT2l63k9Z/xlVpvv9tlc42gc6Pqxp82dfEEzzacSemUY9NZVX638GS4CbJYDupgv+tMkaVk2pAxutpmq0qkrpLx5NYfpV1XBGV6NjKyBX1c/PueXICll9b6XVtVuPtqi/fss1o9GI8xcuoPtdd0EulzsWsGyWtXGtyfe0o6xMDtyZYlr2T5KTdEhIq9XixIkTmD17tnhNLpcjJSUF+fn5Nl9TUFCAadOmWVxLTU3Fvn37AACXL1+GRqNBSkqK+HxISAgSEhKQn5+P0aNHIz8/H+3atRPDCgCkpKRALpfj2LFjuP/++1FQUIB77rlHDCvm+6xduxY3btxAaGhoi5/PnAW1Wq3TgUX+/p8h0/zg8OsMggx/rHsD1TAFimjVNcy5uhSGa7Utv1gZCCEgHAi7C0L77kBEbwhdEoBOcZZ7pggAtFqH6+ZJvSMD0DsyAJOSOsNoFFBUUYNzxdW4UaODTm/6B9TcEyIYjbh67So6duwEuVwOlUIu9sKolHKo5aavZTIZzpdU48z1SpzR3EStzgCjIEClkJv+YVea/rFTKeVQymW4WWdA0Y0aXCqrgVZvxJlrFThzrcIln6+ovBr5F1zyVo47Zcdk1dvcgLs74N0pyQ6/zmAw7Tdk9bOvDAY6RJsezdHVNOyMXFsK2c1SoK4cqK00TRquX8kl01Y0XKurAmorINM3+g+HIJhWh2mb+k+I9O4CIBQCBqkr0gSh20AY//RBq95DoVBALpdzPl0rSRpYysrKYDAYrIZ+wsPDcfas9dwEACguLkZERIRV+eLiYgCARqMRrzVVpri4GB06WE4kVCqVCA0NFV9fXFyMqKgoizLm+xYXF9sVWIxG03DEDz84HjhE97zq9EvftfiuE87iQ+frcQPAjfPOv96LhNY/oKx/NB4hCw0G0MT/bo31DwDhIUD/EAA97Tm4UAEguP5Bt5vCwkKnX9uqn31RmOmhQv1hV+RxrfgzYMZhr9bzikm3bZVSqURcXByTNRGRj5PbsVEmNU/SwBIWFgaFQoGSkhKL6yUlJVa9KGYRERFiT4mt8pGRkeK1jh07WpSJiYkR36O01HLXK71ejxs3boivt3Uf8/dN1e1WcrncYkiJiIiInCNp5FOr1YiNjUVeXp54zWg0Ii8vD0lJSTZfk5iYiMOHD1tcy83NRWJiIgAgKioKkZGRFu9ZVVWFo0ePiu+ZlJSEiooKHD9+XCxz+PBhGI1GcTl1YmIivvvuO+h0Oov73H333XYNBxEREZHrSN5HlZGRge3bt2PXrl34+eefsWTJEtTU1GD8+PEAgAULFohLjQFg6tSp+Oqrr7Bhwwb8/PPPeOONN3D8+HGkp6cDMJ09M3XqVLz99tv44osvcOrUKSxYsAAdO3bEsGHDAAA9e/bEoEGD8Pe//x3Hjh3D999/jxdeeAGjR49Gp06m7d7HjBkDlUqFxYsX46effkJOTg42b96MjIwMD7cQERERSb6sGQC2bt0qbhzXp08fPPvss0hISAAATJkyBV27dsWyZcvE8nv27MGKFSvEjeOeeuopmxvHbd++HRUVFejfvz+ef/553H333WIZ88Zx+/fvFzeOe/bZZ5vcOC4sLAzp6ek2l1oTERGRe3lFYCEiIiJqjuRDQkREREQtYWAhIiIir8fAQkRERF6PgYWIiIi8HgOLD9i2bRvS0tIQFxeHSZMm4dixY1JXyWPWrFmDCRMmICkpCQMHDsScOXOsjn2oq6tDVlYWBgwYgKSkJDz66KNWmwZeuXIFs2bNQkJCAgYOHIhXXnkFer20Bym6y7vvvovo6Gi89NJL4jW2kcm1a9fw5JNPYsCAAYiPj8eYMWMstu4XBAErV65Eamoq4uPjMW3aNJw/f97iPcrLyzF//nwkJyfjnnvuwTPPPIObN11x2KF3MBgMWLFiBdLS0hAfH49hw4Zh9erVaLy+g+1EThGoTfv000+F2NhY4cMPPxR++ukn4dlnnxXuueceobi4WOqqecT06dOFHTt2CKdPnxZOnjwpPPTQQ8Lvfvc74ebNm2KZ5557ThgyZIiQm5srFBYWCpMnTxYefPBB8Xm9Xi888MADwrRp04QffvhBOHDggDBgwADhtddek+IjudXRo0eF++67TxgzZozw4osvitfZRoJQXl4u3HfffcKiRYuEo0ePChcvXhS++uor4cKFC2KZNWvWCP379xc+//xz4eTJk8LDDz8spKWlCbW1tWKZGTNmCH/4wx+EgoIC4dtvvxXuv/9+4W9/+5sUH8kt3n77beE3v/mN8OWXXwqXLl0S9uzZIyQmJgqbNm0Sy7CdyBkMLG3cxIkThaysLPF7g8EgpKamCmvWrJGwVtIpKSkRevfuLXzzzTeCIAhCRUWFEBsbK+zZs0csc+bMGaF3795Cfn6+IAiCcODAASEmJkbQaDRimffff19ITk4W6urqPFp/d6qqqhKGDx8ufP3110J6eroYWNhGJv/4xz+EP/3pT00+bzQahd/+9rfCunXrxGsVFRVCv379hE8++UQQhIZ2O3bsmFjm4MGDQnR0tHD16lX3Vd6DZs2aJTz99NMW1+bOnSvMnz9fEAS2EzmPQ0JtmFarxYkTJ5CSkiJek8vlSElJQX5+voQ1k05lZSUAiMcrHD9+HDqdzqKNevbsiS5duqCgoAAAUFBQgN69e1ucIZWamoqqqiqcOXPGc5V3s6VLl2LIkCEWbQGwjcz279+Pfv364bHHHsPAgQMxduxYbN++XXz+8uXL0Gg0Fu0UEhKChIQE8ectPz8f7dq1Q1xcnFgmJSUFcrm8zQzVJiUl4fDhwzh37hwA0wac33//PQYPHgyA7UTO42nNbVhZWRkMBgPCw8MtroeHh1vN4/AFRqMRL7/8MpKTk9G7d28ApgMtVSoV2rVrZ1E2PDwcGo1GLHPrgZfm781lbneffvopfvjhB3z44YdWz7GNTC5duoR//etfyMjIwMMPP4zCwkK8+OKLUKlUGDdunPg5bf28mef7FBcXo0OHDhbPK5VKhIaGtpl2mjVrFqqqqjBy5EgoFAoYDAY88cQT+MMf/gAAbCdyGgML+YysrCz89NNPeP/996WuilcpKirCSy+9hA0bNsDPz0/q6ngtQRDQr18//O1vfwMA9O3bFz/99BM++OADjBs3TuLaeY89e/bg448/xmuvvYZevXrh5MmTyM7ORseOHdlO1CocEmrDwsLCoFAoUFJSYnG9pKTE6n/Dbd3SpUtx4MABbNq0CXfccYd4PSIiAjqdDhUVFRblS0pKEBkZKZa5dUWM+XtzmdvZiRMnUFJSgvHjx6Nv377o27cvvvnmG2zZsgV9+/ZlG9WLjIxEz549La716NEDV65cEZ8H0OzPW0REBEpLSy2e1+v1uHHjRptpp1dffRWzZs3C6NGjER0djbFjx+Kvf/0r1qxZA4DtRM5jYGnD1Go1YmNjkZeXJ14zGo3Iy8tDUlKShDXzHEEQsHTpUnz++efYtGkTunXrZvF8v379oFKpLNro7NmzuHLlChITEwEAiYmJOH36tMVfsLm5uQgODkavXr088jnc6d5778XHH3+M3bt3i49+/fphzJgx4te+3kYAkJycLM7LMDt//jy6du0KAIiKikJkZKRFO1VVVeHo0aPiz1tSUhIqKipw/Phxsczhw4dhNBoRHx/vgU/hfrW1tZDJZBbXFAqFuKyZ7UTO4pBQG5eRkYGFCxeiX79+iI+Px6ZNm1BTU4Px48dLXTWPyMrKwieffIK33noLQUFB4vh3SEgI/P39ERISggkTJmDZsmUIDQ1FcHAwXnzxRSQlJYn/GKempqJXr15YsGABnnrqKWg0GqxYsQJ/+ctfoFarJfx0rhEcHCzO6TELDAxE+/btxeu+3kYA8Ne//hV/+tOf8M4772DkyJE4duwYtm/fjqVLlwIAZDIZpk6dirfffht33XUXoqKisHLlSnTs2BHDhg0DYJqsPGjQIPz9739HVlYWdDodXnjhBYwePRqdOnWS8uO5zH333Yd33nkHXbp0EYeENm7ciAkTJgBgO5HzeFqzD9i6dSvWr18PjUaDPn364Nlnn0VCQoLU1fKI6Ohom9ezs7PF0FZXV4dly5bh008/hVarRWpqKp5//nmLrudffvkFS5YswTfffIOAgACMGzcO8+fPh1LZNjP/lClTEBMTg8WLFwNgG5l9+eWXeP3113H+/HlERUUhIyMDkydPFp8XBAGrVq3C9u3bUVFRgf79++P555/H3XffLZYpLy/HCy+8gP3790Mul2P48OF49tlnERQUJMVHcrmqqiqsXLkS+/btQ0lJCTp27IjRo0cjMzNTDK9sJ3IGAwsRERF5Pc5hISIiIq/HwEJERERej4GFiIiIvB4DCxEREXk9BhYiIiLyegwsRERE5PUYWIiIiMjrMbAQERGR12NgISIiIq/HwEJERERej4GFiIiIvB4DCxE5zGg0Ys2aNUhLS0N8fDz+8Ic/4LPPPhOfP336NGbOnImkpCSkpKTgqaeeQmlpqfh8VVUV5s+fj8TERKSmpuK9997DlClT8NJLL0nxcYjoNsDAQkQOW7NmDXbv3o2srCx8+umnmDZtGp566il88803qKiowF//+lf07dsXH374IdatW4eSkhLMmzdPfP2yZcuQn5+Pt99+Gxs2bMB3332HEydOSPeBiMjr8bRmInKIVqvFb37zG2zcuBFJSUni9cWLF6O2thY9e/bE999/j/Xr14vPXb16FUOGDMFnn32GyMhI3HvvvVi+fDlGjBgBAKisrMSgQYMwadIkLF682OOfiYi8n1LqChDR7eXChQuoqanB9OnTLa7rdDr06dMHOp0O//vf/yzCjNnFixdRV1cHnU6H+Ph48XpISAjuvvtut9ediG5fDCxE5JDq6moApmGhTp06WTynVqvx97//Hffddx+efPJJq9dGRkbi4sWLHqknEbUtDCxE5JCePXtCrVbjypUr+M1vfmP1fGxsLP7zn/+ga9euUCqt/4qJioqCSqVCYWEhunTpAsA0JHT+/Hncc889bq8/Ed2eGFiIyCHBwcGYPn06srOzIQgC+vfvj8rKShw5cgTBwcH485//jO3bt+Nvf/sbZs6cifbt2+PChQvIycnBiy++iODgYIwdOxavvvoqQkNDER4ejjfeeAMymQwymUzqj0dEXoqBhYgcNm/ePHTo0AFr1qzB5cuXERISgr59++Lhhx9Gp06d8K9//QvLly/HjBkzoNVq0aVLFwwaNAhyuWlh4qJFi/D888/j4YcfRnBwMGbOnImioiL4+flJ/MmIyFtxlRARSa66uhqDBw/GwoULMWnSJKmrQ0ReiD0sRORxP/zwA86ePYv4+HhUVlZi9erVAIChQ4dKXDMi8lYMLEQkiQ0bNuDcuXNQqVSIjY3Ftm3b0KFDB6mrRUReikNCRERE5PW4NT8RERF5PQYWIiIi8noMLEREROT1GFiIiIjI6zGwEBERkddjYCEiIiKvx8BCREREXo+BhYiIiLze/w+GNqidkLz3xQAAAABJRU5ErkJggg==",
-      "text/plain": [
-       "<Figure size 558.875x500 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "sns.displot(\n",
-    "    data=data_unpivoted.groupby([\"sample_index\", \"y\"]).std().reset_index(),\n",
-    "    kind=\"kde\",\n",
-    "    x=\"eeg\",\n",
-    "    hue=\"y\",\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "At this point, it is fairly obvious that standard deviation is going to be a major feature in this analysis."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Creating a GetML Data Model\n",
-    "\n",
-    "Now that we have explored our data, let's do some machine learning. GetML uses a highly sophisticated engine that runs in the background and takes away a lot of hassle in machine learning applications. \n",
-    "\n",
-    "Let's take a look at loading data into your getML project. First, let's learn how we work with data in getML. Data is represented by getML's custom [DataFrame](https://docs.getml.com/latest/api/data/getml.DataFrame.html) that behaves similarly to a pandas DataFrame. However, a [getML.DataFrame](https://docs.getml.com/latest/api/data/getml.DataFrame.html) is a representation of our data inside getML's highly efficient C++ database engine that runs in the background. We can [load data](https://docs.getml.com/latest/user_guide/importing_data/importing_data.html) from various sources such as pandas DataFrames ([`getml.DataFrame.from_pandas`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_pandas.html)), from CSV files ([`getml.DataFrame.from_csv`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_csv.html)), or load from remote databases ([`getml.DataFrame.from_db`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_db.html)) or even S3 buckets ([`getml.DataFrame.from_s3`](https://docs.getml.com/latest/api/data/DataFrame/getml.DataFrame.from_s3.html)).\n",
-    "\n",
-    "Let's create a population DataFrame that contains our main goal: classify a 1s window. This means that we only need a DataFrame that holds the class labels of each window and a unique id, which in this case can just be the `sample_index`."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### The getML DataFrames"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "population_df = pd.DataFrame(\n",
-    "    {\n",
-    "        \"sample_index\": data.index.values,\n",
-    "        \"y\": data.y,\n",
-    "    }\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>sample_index</th>\n",
-       "      <th>y</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>2</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>3</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>4</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11495</th>\n",
-       "      <td>11495</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11496</th>\n",
-       "      <td>11496</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11497</th>\n",
-       "      <td>11497</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11498</th>\n",
-       "      <td>11498</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11499</th>\n",
-       "      <td>11499</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>11500 rows × 2 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       sample_index  y\n",
-       "0                 0  0\n",
-       "1                 1  1\n",
-       "2                 2  0\n",
-       "3                 3  0\n",
-       "4                 4  0\n",
-       "...             ... ..\n",
-       "11495         11495  0\n",
-       "11496         11496  1\n",
-       "11497         11497  0\n",
-       "11498         11498  0\n",
-       "11499         11499  0\n",
-       "\n",
-       "[11500 rows x 2 columns]"
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "population_df"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "A getML.DataFrame can be created from a pandas DataFrame as follows. As this getML.DataFrame is a representation of data internally handled by the engine, we need to specify an internal name. DataFrames are represented by these names in the monitor."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "population = getml.DataFrame.from_pandas(population_df, name=\"population\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th> name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"unused_float\">sample_index</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"unused_float\">           y</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">unused_float</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">unused_float</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">3</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11495</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">11495</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11496</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">11496</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11497</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">11497</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11498</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">11498</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11499</th>\n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">11499</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    11500 rows x 2 columns<br />\n",
-       "    memory usage: 0.18 MB<br />\n",
-       "    name: population<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       " name   sample_index              y\n",
-       " role   unused_float   unused_float\n",
-       "    0              0              0\n",
-       "    1              1              1\n",
-       "    2              2              0\n",
-       "    3              3              0\n",
-       "    4              4              0\n",
-       "                 ...            ...\n",
-       "11495          11495              0\n",
-       "11496          11496              1\n",
-       "11497          11497              0\n",
-       "11498          11498              0\n",
-       "11499          11499              0\n",
-       "\n",
-       "\n",
-       "11500 rows x 2 columns\n",
-       "memory usage: 0.18 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 21,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "population"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As you can see, our data is now stored inside the engine and represented by a getML.DataFrame (data is of course the same). The Python API provides a link to the getML.DataFrame in the monitor, where you can conveniently explore your data.\n",
-    "\n",
-    "Now we need to [annotate our data](https://docs.getml.com/latest/user_guide/annotating_data/annotating_data.html) so the engine knows what to do with it.\n",
-    "\n",
-    "A key aspect of using getML.DataFrame are [roles](https://docs.getml.com/latest/api/getml.data.Roles.html). Every column with relevant data to our data model needs to have a certain role specified. As you can see, both of our columns have the `unused_float` role for now. One of the most important roles is [`getml.data.roles.target`](https://docs.getml.com/latest/api/roles/getml.data.roles.target.html), specifying that the data in this column is our target variable, the value that we want to train our machine learning model on. In our case, the column `y` containing the class label is our target. Let's tell the engine exactly that:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "population.set_role([\"y\"], getml.data.roles.target)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As you may have noticed, our population getML.DataFrame does not contain any actual data, specifically no EEG signals. We utilize one of getML's core strengths here: relational data and time-series.\n",
-    "\n",
-    "Our data can in fact be interpreted as a relational time-series. We have a label for each window with a unique window id (`sample_index`) that stands in relation to its actual data, the corresponding EEG signal of each individual window. Each window has the EEG signal values along with the unique window id (`sample_index`). In other words: we can utilize a very efficient data model by joining a peripheral table containing the EEG values onto the window labels.\n",
-    "\n",
-    "Thus, the next step is to specify the `sample_index` as the join key using `getml.data.roles.join_key` in our population table:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "population.set_role([\"sample_index\"], getml.data.roles.join_key)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th> name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"join_key\">sample_index</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"target\">     y</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\"> role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header join_key\">    join_key</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header target\">target</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">1</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">2</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">3</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">4</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">...</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11495</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11495</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11496</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11496</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11497</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11497</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11498</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11498</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>11499</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align target\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    11500 rows x 2 columns<br />\n",
-       "    memory usage: 0.14 MB<br />\n",
-       "    name: population<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       " name   sample_index        y\n",
-       " role       join_key   target\n",
-       "    0              0        0\n",
-       "    1              1        1\n",
-       "    2              2        0\n",
-       "    3              3        0\n",
-       "    4              4        0\n",
-       "                 ...      ...\n",
-       "11495          11495        0\n",
-       "11496          11496        1\n",
-       "11497          11497        0\n",
-       "11498          11498        0\n",
-       "11499          11499        0\n",
-       "\n",
-       "\n",
-       "11500 rows x 2 columns\n",
-       "memory usage: 0.14 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 24,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "population"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We now create the peripheral table containing the time-series data corresponding to the 1s window labels in the population table just like we did previously with our population table:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "peripheral = getml.DataFrame.from_pandas(data_unpivoted, name=\"peripheral\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now we need to specify the column roles in the peripheral table. This getML.DataFrame contains the `sample_index` as well and we need to set it as our join key, as described above. Subsequently, as this table will contain our actual data in the form of EEG signal values, we specify the role of this column as numerical ([`getml.data.roles.numerical`](https://docs.getml.com/latest/api/roles/getml.data.roles.numerical.html)), something we can train our machine learning model on:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "peripheral.set_role([\"sample_index\"], getml.data.roles.join_key)\n",
-    "peripheral.set_role([\"eeg\"], getml.data.roles.numerical)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  table {\n",
-       "    font-family: Helvetica, sans-serif;\n",
-       "  }\n",
-       "\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right !important;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.sub-header {\n",
-       "    font-weight: normal;\n",
-       "    font-style: italic;\n",
-       "  }\n",
-       "  .join_key,\n",
-       "  .numerical,\n",
-       "  .target,\n",
-       "  .unused_float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "\n",
-       "  .char-align {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  span.left {\n",
-       "    text-align: right;\n",
-       "    width: 3em;\n",
-       "  }\n",
-       "  span.right {\n",
-       "    float: right;\n",
-       "    text-align: left;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th>   name</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"join_key\">sample_index</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"numerical\">      eeg</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"unused_float\">  time_index</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"unused_float\">           y</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header\">   role</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header join_key\">    join_key</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header numerical\">numerical</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">unused_float</th>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            <th class=\"sub-header unused_float\">unused_float</th>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">135</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">1</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">190</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">229</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">3</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">223</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">0</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">192</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">5</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">...</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">...</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2046995</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">5</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">174</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2046996</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">4</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">175</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2046997</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">-2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">176</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2046998</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">2</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">177</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2046999</th>\n",
-       "        \n",
-       "          \n",
-       "            <td class=\"join_key\">11499</td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align numerical\">\n",
-       "              <span class=\"left\">20</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">178</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "          \n",
-       "            \n",
-       "              \n",
-       "              \n",
-       "          \n",
-       "            <td class=\"char-align unused_float\">\n",
-       "              <span class=\"left\">0</span\n",
-       "              ><span class=\"right\" style=\"width: 0ch\"\n",
-       "                >&nbsp;</span\n",
-       "              >\n",
-       "            </td>\n",
-       "          \n",
-       "        \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "\n",
-       "  <p>\n",
-       "    2047000 rows x 4 columns<br />\n",
-       "    memory usage: 57.32 MB<br />\n",
-       "    name: peripheral<br />\n",
-       "    type: getml.DataFrame<br />\n",
-       "    \n",
-       "  </p>\n"
-      ],
-      "text/plain": [
-       "   name   sample_index         eeg     time_index              y\n",
-       "   role       join_key   numerical   unused_float   unused_float\n",
-       "      0              0        135               1              0\n",
-       "      1              0        190               2              0\n",
-       "      2              0        229               3              0\n",
-       "      3              0        223               4              0\n",
-       "      4              0        192               5              0\n",
-       "                   ...         ...            ...            ...\n",
-       "2046995          11499          5             174              0\n",
-       "2046996          11499          4             175              0\n",
-       "2046997          11499         -2             176              0\n",
-       "2046998          11499          2             177              0\n",
-       "2046999          11499         20             178              0\n",
-       "\n",
-       "\n",
-       "2047000 rows x 4 columns\n",
-       "memory usage: 57.32 MB\n",
-       "type: getml.DataFrame"
-      ]
-     },
-     "execution_count": 27,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "peripheral"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As you may have noticed, there are still `unused_float` columns left. This data is present, but we do not use or need it in our machine learning efforts. This unused data is not considered and can just be ignored."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### The getML Data Model\n",
-    "\n",
-    "Now that we have our data efficiently stored in getML.DataFrame, we continue to construct our data model.\n",
-    "\n",
-    "This is very easily done by using one of getML's many [DataModels](https://docs.getml.com/latest/user_guide/data_model/data_model.html). We put our time-series data in a relational context and can utilze for example a simple [StarSchema](https://docs.getml.com/latest/api/getml.data.StarSchema.html) data model to accomplish this. Easily put, we see our windows (the time-series data) as splits into many individual samples that are joined onto the window labels. This way, we are effectively thinking of time series as relational data: we are identifying relevant information from our data and aggragate it into a single label. In fact, what we are doing is effectively a self join, because we are joining a table to itself. This allows for very efficient calculation.\n",
-    "\n",
-    "\n",
-    "First, we define a random data [split](https://docs.getml.com/latest/api/getml.data.split.html):"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 28,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "train, test = 0.9, 0.1\n",
-    "\n",
-    "split = getml.data.split.random(seed=5849, train=train, test=test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Second, we create our data model. We create a StarSchema containing our population getML.DataFrame as the population table and specify the split of our dataset into train and test set. We then [join](https://docs.getml.com/latest/api/StarSchema/getml.data.StarSchema.join.html) our peripheral table to our time series on the join key, in this case `sample_index`:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 29,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "time_series = getml.data.StarSchema(population=population, split=split)\n",
-    "\n",
-    "time_series.join(\n",
-    "    peripheral,\n",
-    "    on=\"sample_index\",\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "getML provides a convenient view to our data model. We can look at a diagram representation of our data model with table names and specific joins, as well as the staging tables and statistics about the underlying data container."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<span style='font-size: 1.2rem; font-weight: 500;'>data model</span>\n",
-       "            <div style='margin-top: 15px; margin-bottom: 5px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>diagram</div>\n",
-       "            <div style=\"height:100px;width:660px;position:relative;\"><svg height=\"90\" width=\"650\"><rect y=\"0\" x=\"0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"75.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">peripheral</text><rect x=\"51\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"67.0\" y1=\"10\" x2=\"67.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"83.0\" y1=\"10\" x2=\"83.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"26.0\" x2=\"99\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"51\" y1=\"42.0\" x2=\"99\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><rect y=\"0\" x=\"500\" rx=\"10\" ry=\"10\" width=\"150\" height=\"90\" style=\"fill:#6829c2;stroke-width:0;\" /><text y=\"73.8\" x=\"575.0\" dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\">population</text><rect x=\"551\" y=\"10\" rx=\"4\" ry=\"4\" width=\"48\" height=\"48\" style=\" fill:#6829c2;stroke:#ffffff;stroke-width:3;\" /><line x1=\"567.0\" y1=\"10\" x2=\"567.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"583.0\" y1=\"10\" x2=\"583.0\" y2=\"58\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"26.0\" x2=\"599\" y2=\"26.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"551\" y1=\"42.0\" x2=\"599\" y2=\"42.0\" style=\"stroke:white;stroke-width:3\" /><line x1=\"150\" y1=\"43.0\" x2=\"490\" y2=\"43.0\" style=\"stroke:#808080;;stroke-width:4\" /><polygon points=\"500, 43.0 490, 37.0 490, 49.0 \" style=\"fill:#808080;;stroke-width:0;\" /><rect y=\"10.0\" x=\"249.0\" rx=\"10\" ry=\"10\" width=\"150\" height=\"70\" style=\"fill:#6829c2;stroke-width:0;\" /><text dominant-baseline=\"middle\" text-anchor=\"middle\" fill=\"white\"><tspan y=\"45.0\" x=\"324.0\" font-size=\"7pt\" >sample_index = sample_index</tspan></text></svg></div>\n",
-       "            </div>\n",
-       "\n",
-       "            <div style='margin-top: 15px;'>\n",
-       "            <div style='margin-bottom: 10px; font-size: 1rem;'>staging</div>\n",
-       "            <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">data frames</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">staging table              </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">POPULATION__STAGING_TABLE_1</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">peripheral</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">PERIPHERAL__STAGING_TABLE_2</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "            </div>\n",
-       "            \n",
-       "<span style='font-size: 1.2rem; font-weight: 500;'>container</span>\n",
-       "<div style='margin-top: 15px;'>\n",
-       "<div style='float: left; margin-right: 50px;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>population</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">subset</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name      </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> rows</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">1086</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">population</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">10414</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">View</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "<div style='float: left;'>\n",
-       "<div style='margin-bottom: 10px; font-size: 1rem;'>peripheral</div>\n",
-       "    <style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name      </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\">   rows</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">type     </th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">peripheral</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"int\">2047000</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">DataFrame</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "data model\n",
-       "\n",
-       "  population:\n",
-       "    columns:\n",
-       "    - sample_index: join_key\n",
-       "    - y: target\n",
-       "\n",
-       "    joins:\n",
-       "    - right: 'peripheral'\n",
-       "      on: (population.sample_index, peripheral.sample_index)\n",
-       "      relationship: 'many-to-many'\n",
-       "      lagged_targets: False\n",
-       "\n",
-       "  peripheral:\n",
-       "    columns:\n",
-       "    - sample_index: join_key\n",
-       "    - eeg: numerical\n",
-       "    - time_index: unused_float\n",
-       "    - y: unused_float\n",
-       "\n",
-       "\n",
-       "container\n",
-       "\n",
-       "  population\n",
-       "      subset   name          rows   type\n",
-       "  0   test     population    1086   View\n",
-       "  1   train    population   10414   View\n",
-       "\n",
-       "  peripheral\n",
-       "      name            rows   type     \n",
-       "  0   peripheral   2047000   DataFrame"
-      ]
-     },
-     "execution_count": 30,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "time_series"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This is an overview of your data model in the getML engine. At the top you can see a visual representation in the form of a diagram. Here you can easily see how your data and the specific joins is structured. Next you are presented the so called staging tables. This is a list of the relevant data frames and staging table names. At last, you can see an overview of all the data [containers](https://docs.getml.com/latest/api/getml.data.Container.html). This includes the split in train and test set of your population table as well as the peripheral tables.\n",
-    "\n",
-    "In this simple example, the diagram consists of a single join of the peripheral table onto the population table via the `sample_index` as a join key. The population table is split into 90% train and 10% test set. The peripheral talbe contains all the EEG signal values and has over 2 million rows."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### The getML machine learning pipeline\n",
-    "\n",
-    "Complex machine learning models are represented by getML [pipelines](https://docs.getml.com/latest/api/pipeline/getml.pipeline.Pipeline.html). A pipeline contains the data model (including complex data relations), data [preprocessors](https://docs.getml.com/latest/api_reference/preprocessors.html), [feature learners](https://docs.getml.com/latest/api_reference/feature_learning.html), [predictors](https://docs.getml.com/latest/api_reference/predictors.html) and so on.\n",
-    "\n",
-    "In our approach, we will use getML's very own [FastProp](https://docs.getml.com/latest/api/getml.feature_learning.FastProp.html) automatic feature learner for [feature engineering](https://docs.getml.com/latest/user_guide/feature_engineering/feature_engineering.html). We specify a loss function suitable for classification. As we are only dealing with a univariate time-series, we want to use all possible aggregation functions.\n",
-    "\n",
-    "We use the highly efficient [XGBoost](https://docs.getml.com/latest/api/getml.predictors.XGBoostClassifier.html) classifier algorithm as a [predictor](https://docs.getml.com/latest/user_guide/predicting/predicting.html)."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 31,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "feature_learner = getml.feature_learning.FastProp(\n",
-    "    loss_function=getml.feature_learning.loss_functions.CrossEntropyLoss,\n",
-    "    aggregation=getml.feature_learning.FastProp.agg_sets.All,\n",
-    ")\n",
-    "\n",
-    "predictor = getml.predictors.XGBoostClassifier()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now that we have our data model and machine learning components defined in just a few lines of code, we declare our machine learning pipeline as simple as the following. For convenience, we specify some free to choose tags. These are shown in the monitor and can be used to efficiently and easily distinguish different pipelines and their performance."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "pipe = getml.pipeline.Pipeline(\n",
-    "    data_model=time_series.data_model,\n",
-    "    tags=[\"FastProp+AggAll\", \"XGBoost\", f\"split={train}/{test}\"],\n",
-    "    feature_learners=[feature_learner],\n",
-    "    predictors=[predictor],\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now all we need to do is train our model:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Trying 25 features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "XGBoost: Training as predictor... 100% |██████████| [elapsed: 00:01, remaining: 00:00]          \n",
-      "\n",
-      "Trained pipeline.\n",
-      "Time taken: 0h:0m:1.434615\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre>Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='CrossEntropyLoss',\n",
-       "         peripheral=['peripheral'],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['FastProp+AggAll', 'XGBoost', 'split=0.9/0.1', 'container-Mq37qh'])</pre>"
-      ],
-      "text/plain": [
-       "Pipeline(data_model='population',\n",
-       "         feature_learners=['FastProp'],\n",
-       "         feature_selectors=[],\n",
-       "         include_categorical=False,\n",
-       "         loss_function='CrossEntropyLoss',\n",
-       "         peripheral=['peripheral'],\n",
-       "         predictors=['XGBoostClassifier'],\n",
-       "         preprocessors=[],\n",
-       "         share_selected_features=0.5,\n",
-       "         tags=['FastProp+AggAll', 'XGBoost', 'split=0.9/0.1', 'container-Mq37qh'])"
-      ]
-     },
-     "execution_count": 33,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe.fit(time_series.train, check=False)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This is all it takes. FastProp found features in just 1 second and XGBoost trained our model in just 2 seconds. The whole process took less than 4 seconds!"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Model Evaluation"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Now, let's look at how well our model performs. Again, getML does everything for you. We [score](https://docs.getml.com/latest/api/pipeline/Pipeline/getml.pipeline.Pipeline.score.html) our pipeline on the test set:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 34,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Staging... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "Preprocessing... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "FastProp: Building features... 100% |██████████| [elapsed: 00:00, remaining: 00:00]          \n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\"> </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"datetime\">date time          </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">set used</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">accuracy</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">    auc</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">cross entropy</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-08-26 15:14:26</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">train</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.98</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9974</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.05586</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"datetime\">2024-08-26 15:14:26</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">test</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.9696</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.994</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0793</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "    date time             set used   target   accuracy       auc   cross entropy\n",
-       "0   2024-08-26 15:14:26   train      y          0.98      0.9974         0.05586\n",
-       "1   2024-08-26 15:14:26   test       y          0.9696    0.994          0.0793 "
-      ]
-     },
-     "execution_count": 34,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe.score(time_series.test)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Let's have a look at some key [machine learning metrics](https://docs.getml.com/latest/api/pipeline/getml.pipeline.Scores.html): Accuracy and Area Under Curve (AUC):"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 35,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Accuracy: 96.96%, AUC: 0.9940\n"
-     ]
-    }
-   ],
-   "source": [
-    "print(f\"Accuracy: {pipe.scores.accuracy*100:.2f}%, AUC: {pipe.scores.auc:.4f}\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "There are several other, more complex metrics to understand the performance of a machine learning model. The most prominent being Receiver Operating Characteristic (ROC) curve, precision-recall curve and lift curve. getML has these already calculated for you:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 36,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1200x400 with 3 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "# Refers to the data from the last time we called .score(...)\n",
-    "fpr, tpr = pipe.plots.roc_curve()\n",
-    "recall, precision = pipe.plots.precision_recall_curve()\n",
-    "proportion, lift = pipe.plots.lift_curve()\n",
-    "\n",
-    "fig, ax = plt.subplots(ncols=3, figsize=(12, 4))\n",
-    "\n",
-    "ax[0].plot(fpr, tpr, color=\"#6829c2\")\n",
-    "ax[0].set_title(\"receiver operating characteristic (ROC)\")\n",
-    "ax[0].set_xlabel(\"false positive rate\")\n",
-    "ax[0].set_ylabel(\"true positive rate\")\n",
-    "\n",
-    "ax[1].plot(recall, precision, color=\"#6829c2\")\n",
-    "ax[1].set_title(\"precision-recall curve\")\n",
-    "ax[1].set_xlabel(\"recall (true positive rate)\")\n",
-    "ax[1].set_ylabel(\"precision\")\n",
-    "\n",
-    "ax[2].plot(proportion, lift, color=\"#6829c2\")\n",
-    "ax[2].set_title(\"lift curve\")\n",
-    "ax[2].set_xlabel(\"proportion\")\n",
-    "ax[2].set_ylabel(\"lift\")\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The Receiver-Operator-Characteristic (ROC) curve is a diagram that shows the diagnostic ability of a binary classifier for all classification thresholds. It shows the tradeoff between sensitivity (true positive rate or TPR) and specificity (1 - FPR or false positive rate). Easily put, the closer the curve to the top left (meaning the larger the area under curve or AUC) , the more accurate the classifier. A 45° diagonal would be a random classifier.\n",
-    "\n",
-    "Much like the ROC curve, a precision-recall curve (PR curve) is used to evaluate the performance of a binary classifier. It is often used when dealing with heavily imbalanced classes. It is desired that your machine learning model has both high precision and high recall. However, we often end up with a trade-off between the two. Similar to a ROC curve, the higher the area und the curve the better performing our binary classifier.\n",
-    "\n",
-    "The Lift curve shows the relation between the number of instances which were predicted positive and those that are indeed positive and thus, like ROC and PR curves, measures the effectiveness of a chosen classifier against a random classifier. In our example, the patients with the highest probability of having an epileptic seizure appear on the left of the Lift curve along with high Lift scores. This point is called the Maximum Lift Point: the higher this point, the better our model performs. Also, it is generally considered that the longer the flat part on the right of the Lift curve, the more reliable the model is.\n",
-    "\n",
-    "All three performance diagrams measure the performance of a binary classifier against a random classifier. As a rule of thumb, the higher ROC and PR curves the better, while a Lift curve is desired to be high in the left and preferably flat on the right."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Studying the Features\n",
-    "\n",
-    "Finally, we can have a look at the features our relational learning algorithm has extracted. We can view them conveniently in the getML monitor under the respective pipeline, or print them directly in Python:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 37,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<style>\n",
-       "  th {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  td {\n",
-       "    text-align: left !important;\n",
-       "  }\n",
-       "  th:nth-child(1) {\n",
-       "    text-align: right;\n",
-       "    border-right: 1px solid LightGray;\n",
-       "  }\n",
-       "  th.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.float {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  th.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "  td.int {\n",
-       "    text-align: right !important;\n",
-       "  }\n",
-       "</style>\n",
-       "\n",
-       "<table class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr>\n",
-       "      \n",
-       "        \n",
-       "          <th class=\"int\">  </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">target</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"str\">name        </th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">correlation</th>\n",
-       "        \n",
-       "      \n",
-       "        \n",
-       "          <th class=\"float\">importance</th>\n",
-       "        \n",
-       "      \n",
-       "    </tr>\n",
-       "    \n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>0</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_1</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0358</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0087</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>1</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_2</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.7295</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.018</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>2</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_3</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">-0.7295</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>3</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_4</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.7646</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.008</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>4</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_5</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0516</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0041</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th></th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">...</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">...</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">...</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">...</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>20</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_21</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.7634</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0012</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>21</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_22</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.7615</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0017</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>22</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_23</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0186</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0189</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>23</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_24</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">-0.0433</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0026</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "      <tr>\n",
-       "        <th>24</th>\n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">y</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"str\">feature_1_25</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0</td>\n",
-       "            \n",
-       "          \n",
-       "            \n",
-       "              <td class=\"float\">0.0</td>\n",
-       "            \n",
-       "          \n",
-       "      </tr>\n",
-       "    \n",
-       "  </tbody>\n",
-       "</table>"
-      ],
-      "text/plain": [
-       "     target   name           correlation   importance\n",
-       " 0   y        feature_1_1         0.0358       0.0087\n",
-       " 1   y        feature_1_2         0.7295       0.018 \n",
-       " 2   y        feature_1_3        -0.7295       0.0   \n",
-       " 3   y        feature_1_4         0.7646       0.008 \n",
-       " 4   y        feature_1_5         0.0516       0.0041\n",
-       "     ...      ...                    ...          ...\n",
-       "20   y        feature_1_21        0.7634       0.0012\n",
-       "21   y        feature_1_22        0.7615       0.0017\n",
-       "22   y        feature_1_23        0.0186       0.0189\n",
-       "23   y        feature_1_24       -0.0433       0.0026\n",
-       "24   y        feature_1_25        0.0          0.0   "
-      ]
-     },
-     "execution_count": 37,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe.features"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We can look at feature correlations:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 38,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 800x400 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "names, correlations = pipe.features.correlations()\n",
-    "\n",
-    "plt.subplots(figsize=(8, 4))\n",
-    "\n",
-    "plt.bar(names, correlations, color=\"#6829c2\")\n",
-    "\n",
-    "plt.title(\"feature correlations\")\n",
-    "plt.xlabel(\"features\")\n",
-    "plt.ylabel(\"correlations\")\n",
-    "plt.xticks(rotation=\"vertical\")\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Or we can have a look at the feature importances. This is particularly interesting if we want to understand what generated features are most important for our machine learning model.\n",
-    "\n",
-    "The feature importance is calculated by XGBoost based on the improvement of the optimizing criterium at each split in the decision tree and is normalized to 100%."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 39,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 800x400 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "names, importances = pipe.features.importances()\n",
-    "\n",
-    "plt.subplots(figsize=(8, 4))\n",
-    "\n",
-    "plt.bar(names, importances, color=\"#6829c2\")\n",
-    "\n",
-    "plt.title(\"feature importances\")\n",
-    "plt.xlabel(\"features\")\n",
-    "plt.ylabel(\"importance\")\n",
-    "plt.xticks(rotation=\"vertical\")\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This is intriguing! It seems that one particular feature stands out in its importance for classification predictions.\n",
-    "\n",
-    "As we already discussed above, we view our time series data from a relation data point of view. In fact, relational learning is one of getML's core strengths. Particularly, getML is able to transpile features into database queries that can be used in production database environments without the need of any other software component.\n",
-    "\n",
-    "Let's have a look at the SQL code of our most important feature:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 40,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/markdown": [
-       "```sql\n",
-       "DROP TABLE IF EXISTS \"FEATURE_1_7\";\n",
-       "\n",
-       "CREATE TABLE \"FEATURE_1_7\" AS\n",
-       "SELECT STDDEV( t2.\"eeg\" ) AS \"feature_1_7\",\n",
-       "       t1.rowid AS rownum\n",
-       "FROM \"POPULATION__STAGING_TABLE_1\" t1\n",
-       "INNER JOIN \"PERIPHERAL__STAGING_TABLE_2\" t2\n",
-       "ON t1.\"sample_index\" = t2.\"sample_index\"\n",
-       "GROUP BY t1.rowid;\n",
-       "```"
-      ],
-      "text/plain": [
-       "'DROP TABLE IF EXISTS \"FEATURE_1_7\";\\n\\nCREATE TABLE \"FEATURE_1_7\" AS\\nSELECT STDDEV( t2.\"eeg\" ) AS \"feature_1_7\",\\n       t1.rowid AS rownum\\nFROM \"POPULATION__STAGING_TABLE_1\" t1\\nINNER JOIN \"PERIPHERAL__STAGING_TABLE_2\" t2\\nON t1.\"sample_index\" = t2.\"sample_index\"\\nGROUP BY t1.rowid;'"
-      ]
-     },
-     "execution_count": 40,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pipe.features.to_sql()[names[0]]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As you can see, the most important feature in seizure recognition from EEG signals seems to be the standard deviation. Just like we guessed in the beginning of this notebook.\n",
-    "\n",
-    "However, relevant features are by far not always so obvious as in this particular example dataset. In fact, most of the time feature engineering takes a lot of effort and domain knowledge from domain experts. As we discussed above, manual feature engineering is a cumbersome, time consuming and error prone process.\n",
-    "\n",
-    "Novel machine learning libraries like getML with automatic feature learning, flexible data models and machine learning pipelines, all wrapped inside an easy to use Python API, backed by an efficient and fast C++ backend make this task a lot easier and way more efficient for data scientists. "
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.10.12"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
+    "nbformat": 4,
+    "nbformat_minor": 4
+}
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
index 1e3f39f..667affe 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,7 +1,7 @@
 jupyterlab==4.1.1
 getml==1.4.0
-featuretools==1.28.0
-tsfresh==0.20.2
+featuretools==1.31.0
+tsfresh==0.20.3
 pyspark==3.5.0
 seaborn==0.13.2
 ipywidgets==8.1.2