5.8. Path Embedding Segmentation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b4386dca",
   "metadata": {},
   "source": [
    "# <font color='CC3D3D'> 0.5_8 Path_Embedding_Segmentation\n",
    "    \n",
    "- 대출 신청을 한 유저에 관한 경로 정보 path_applied_user.csv, 하지 않은 유저에 관한 경로 정보 path_no_applied_user.csv, 대출 신청 여부가 드러나 있지 않은 유저에 관한 경로 정보 path_null_applied_user.csv를 취합\n",
    "- NLP 모델 Doc2Vec을 사용하여 경로 정보를 같은 길이로 임베딩한 뒤,\n",
    "- 거리 기반의 KMeans 군집화 알고리즘을 사용하여 경로가 비슷한 6개의 그룹으로 구분\n",
    "- 구분 데이터 <span style=\"color:blue\"> **path_group.csv** </span> 생성"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47860ce2",
   "metadata": {},
   "source": [
    "# Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ed1d151a",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:43:26.054392Z",
     "iopub.status.busy": "2022-10-06T10:43:26.053871Z",
     "iopub.status.idle": "2022-10-06T10:43:27.988061Z",
     "shell.execute_reply": "2022-10-06T10:43:27.987054Z"
    },
    "id": "ed1d151a",
    "outputId": "cd0c1ed3-f68c-4822-8f99-8d25ee196c16",
    "papermill": {
     "duration": 1.943748,
     "end_time": "2022-10-06T10:43:27.990405",
     "exception": false,
     "start_time": "2022-10-06T10:43:26.046657",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from tqdm import tqdm\n",
    "import pickle\n",
    "from IPython.display import display\n",
    "from numpy.typing import ArrayLike\n",
    "\n",
    "import nltk\n",
    "import gensim\n",
    "from gensim.models.doc2vec import Doc2Vec, TaggedDocument\n",
    "from nltk.tokenize import word_tokenize\n",
    "import gc\n",
    "import warnings\n",
    "\n",
    "from sklearn.cluster import KMeans, AgglomerativeClustering\n",
    "from sklearn.metrics import silhouette_score\n",
    "\n",
    "pd.set_option('display.max_columns', None)\n",
    "nltk.download('punkt')\n",
    "warnings.filterwarnings('ignore')\n",
    "RANDOM_STATE = 52947"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6602db5a",
   "metadata": {},
   "source": [
    "# Def"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "efe909ed",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T10:43:28.001546Z",
     "iopub.status.busy": "2022-10-06T10:43:28.001126Z",
     "iopub.status.idle": "2022-10-06T10:43:28.006143Z",
     "shell.execute_reply": "2022-10-06T10:43:28.005238Z"
    },
    "id": "efe909ed",
    "papermill": {
     "duration": 0.012774,
     "end_time": "2022-10-06T10:43:28.008025",
     "exception": false,
     "start_time": "2022-10-06T10:43:27.995251",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def viewinfo(df:pd.DataFrame) -> None:\n",
    "    print(df.shape)\n",
    "    print()\n",
    "    df.info()\n",
    "    print()\n",
    "    display(df.head(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ZlbPHDdHczB_",
   "metadata": {
    "id": "ZlbPHDdHczB_"
   },
   "outputs": [],
   "source": [
    "def minmax(series:ArrayLike) -> ArrayLike:\n",
    "    maximum, minimum = series.max(), series.min()\n",
    "    return (series-minimum) / (maximum-minimum)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2397a268",
   "metadata": {},
   "source": [
    "# Data Load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "88a7d38d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 595
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:43:28.019812Z",
     "iopub.status.busy": "2022-10-06T10:43:28.019419Z",
     "iopub.status.idle": "2022-10-06T10:43:36.181449Z",
     "shell.execute_reply": "2022-10-06T10:43:36.180043Z"
    },
    "id": "88a7d38d",
    "outputId": "ced9db98-be82-4ee8-e6cc-f572968d67de",
    "papermill": {
     "duration": 8.170238,
     "end_time": "2022-10-06T10:43:36.183288",
     "exception": false,
     "start_time": "2022-10-06T10:43:28.013050",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(679996, 3)\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 679996 entries, 0 to 323290\n",
      "Data columns (total 3 columns):\n",
      " #   Column          Non-Null Count   Dtype \n",
      "---  ------          --------------   ----- \n",
      " 0   application_id  679996 non-null  int64 \n",
      " 1   path            679996 non-null  object\n",
      " 2   date_cd         679996 non-null  object\n",
      "dtypes: int64(1), object(2)\n",
      "memory usage: 20.8+ MB\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-58957a9d-c216-4424-97aa-9aea973b4151\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>application_id</th>\n",
       "      <th>path</th>\n",
       "      <th>date_cd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>871260</td>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>94017</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1482636</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1345169</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>75852</td>\n",
       "      <td>('OpenApp', 'Login', 'UseLoanManage', 'GetCred...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1563807</td>\n",
       "      <td>('GetCreditInfo', 'UseLoanManage', 'GetCreditI...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1803041</td>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>260091</td>\n",
       "      <td>('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2089285</td>\n",
       "      <td>('Login',)</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>554652</td>\n",
       "      <td>('StartLoanApply', 'ViewLoanApplyIntro', 'Open...</td>\n",
       "      <td>2022-03-25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-58957a9d-c216-4424-97aa-9aea973b4151')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-58957a9d-c216-4424-97aa-9aea973b4151 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-58957a9d-c216-4424-97aa-9aea973b4151');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   application_id                                               path  \\\n",
       "0          871260  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "1           94017  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "2         1482636  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "3         1345169  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "4           75852  ('OpenApp', 'Login', 'UseLoanManage', 'GetCred...   \n",
       "5         1563807  ('GetCreditInfo', 'UseLoanManage', 'GetCreditI...   \n",
       "6         1803041  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "7          260091  ('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...   \n",
       "8         2089285                                         ('Login',)   \n",
       "9          554652  ('StartLoanApply', 'ViewLoanApplyIntro', 'Open...   \n",
       "\n",
       "      date_cd  \n",
       "0  2022-03-25  \n",
       "1  2022-03-25  \n",
       "2  2022-03-25  \n",
       "3  2022-03-25  \n",
       "4  2022-03-25  \n",
       "5  2022-03-25  \n",
       "6  2022-03-25  \n",
       "7  2022-03-25  \n",
       "8  2022-03-25  \n",
       "9  2022-03-25  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "applied_path = pd.read_csv(\"../Data/path_applied_user.csv\", engine='c', low_memory=True)\n",
    "non_applied_path = pd.read_csv(\"../Data/path_no_applied_user.csv\", engine='c', low_memory=True)\n",
    "null_applied_path = pd.read_csv(f\"../Data/path_null_applied_user.csv\", engine='c', low_memory=True)\n",
    "user = pd.read_csv(\"../Data/2022빅콘테스트_데이터분석리그_데이터분석분야_퓨처스부문_데이터셋_220908/user_spec.csv\", engine='c',\n",
    "                   dtype={'application_id': 'uint32', 'user_id': 'uint32'},\n",
    "                   low_memory=True,\n",
    "                   parse_dates=['insert_time'],\n",
    "                   usecols=['application_id', 'user_id', 'insert_time'])\n",
    "\n",
    "paths = pd.concat([applied_path, non_applied_path], axis=0)\n",
    "viewinfo(paths)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f8047513",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T10:43:36.194650Z",
     "iopub.status.busy": "2022-10-06T10:43:36.194343Z",
     "iopub.status.idle": "2022-10-06T10:44:01.382566Z",
     "shell.execute_reply": "2022-10-06T10:44:01.381553Z"
    },
    "id": "f8047513",
    "papermill": {
     "duration": 25.196355,
     "end_time": "2022-10-06T10:44:01.384861",
     "exception": false,
     "start_time": "2022-10-06T10:43:36.188506",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "app_user_id = {user['application_id'].iloc[i]: user['user_id'].iloc[i] for i in range(user.shape[0])}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "959d8fb1",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 613
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:44:01.396585Z",
     "iopub.status.busy": "2022-10-06T10:44:01.396309Z",
     "iopub.status.idle": "2022-10-06T10:44:02.362315Z",
     "shell.execute_reply": "2022-10-06T10:44:02.361177Z"
    },
    "id": "959d8fb1",
    "outputId": "c81a2477-609c-4c02-a251-041b15e3a3de",
    "papermill": {
     "duration": 0.975064,
     "end_time": "2022-10-06T10:44:02.365316",
     "exception": false,
     "start_time": "2022-10-06T10:44:01.390252",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1740729, 4)\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1740729 entries, 0 to 1060732\n",
      "Data columns (total 4 columns):\n",
      " #   Column          Dtype  \n",
      "---  ------          -----  \n",
      " 0   application_id  float64\n",
      " 1   path            object \n",
      " 2   date_cd         object \n",
      " 3   user_id         int64  \n",
      "dtypes: float64(1), int64(1), object(2)\n",
      "memory usage: 66.4+ MB\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-51e32557-fc14-4e25-b5cf-6da97ddfaf38\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>application_id</th>\n",
       "      <th>path</th>\n",
       "      <th>date_cd</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>871260.0</td>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>72878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>94017.0</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>618857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1482636.0</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>618857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1345169.0</td>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>496477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>75852.0</td>\n",
       "      <td>('OpenApp', 'Login', 'UseLoanManage', 'GetCred...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>217308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1563807.0</td>\n",
       "      <td>('GetCreditInfo', 'UseLoanManage', 'GetCreditI...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>27667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1803041.0</td>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>831139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>260091.0</td>\n",
       "      <td>('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>229879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2089285.0</td>\n",
       "      <td>('Login',)</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>193072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>554652.0</td>\n",
       "      <td>('StartLoanApply', 'ViewLoanApplyIntro', 'Open...</td>\n",
       "      <td>2022-03-25</td>\n",
       "      <td>102002</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-51e32557-fc14-4e25-b5cf-6da97ddfaf38')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-51e32557-fc14-4e25-b5cf-6da97ddfaf38 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-51e32557-fc14-4e25-b5cf-6da97ddfaf38');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   application_id                                               path  \\\n",
       "0        871260.0  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "1         94017.0  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "2       1482636.0  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "3       1345169.0  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "4         75852.0  ('OpenApp', 'Login', 'UseLoanManage', 'GetCred...   \n",
       "5       1563807.0  ('GetCreditInfo', 'UseLoanManage', 'GetCreditI...   \n",
       "6       1803041.0  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "7        260091.0  ('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...   \n",
       "8       2089285.0                                         ('Login',)   \n",
       "9        554652.0  ('StartLoanApply', 'ViewLoanApplyIntro', 'Open...   \n",
       "\n",
       "      date_cd  user_id  \n",
       "0  2022-03-25    72878  \n",
       "1  2022-03-25   618857  \n",
       "2  2022-03-25   618857  \n",
       "3  2022-03-25   496477  \n",
       "4  2022-03-25   217308  \n",
       "5  2022-03-25    27667  \n",
       "6  2022-03-25   831139  \n",
       "7  2022-03-25   229879  \n",
       "8  2022-03-25   193072  \n",
       "9  2022-03-25   102002  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "paths['user_id'] = paths['application_id'].map(app_user_id)\n",
    "paths = pd.concat([paths, null_applied_path], axis=0)\n",
    "viewinfo(paths)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "JsNAc_kBddci",
   "metadata": {
    "id": "JsNAc_kBddci"
   },
   "outputs": [],
   "source": [
    "unique_paths = pd.DataFrame({'path': paths['path'].unique()})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "hrf14FnRdxI-",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 363
    },
    "id": "hrf14FnRdxI-",
    "outputId": "9b7b23f2-b264-4c2e-ecc5-cca3b805329b"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-03077b4c-91b8-495c-9ee3-1fc5cf8c9857\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>path</th>\n",
       "      <th>processed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>('OpenApp', 'Login', 'UseLoanManage', 'GetCred...</td>\n",
       "      <td>open app, login, use loan manage, get credit i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>('GetCreditInfo', 'UseLoanManage', 'GetCreditI...</td>\n",
       "      <td>get credit information, use loan manage, get c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...</td>\n",
       "      <td>end loan apply, view loan apply intro, complet...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>('Login',)</td>\n",
       "      <td>login</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>('StartLoanApply', 'ViewLoanApplyIntro', 'Open...</td>\n",
       "      <td>start loan apply, view loan apply intro, open ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-03077b4c-91b8-495c-9ee3-1fc5cf8c9857')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-03077b4c-91b8-495c-9ee3-1fc5cf8c9857 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-03077b4c-91b8-495c-9ee3-1fc5cf8c9857');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "                                                path  \\\n",
       "0  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "1  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "2  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "3  ('OpenApp', 'Login', 'UseLoanManage', 'GetCred...   \n",
       "4  ('GetCreditInfo', 'UseLoanManage', 'GetCreditI...   \n",
       "5  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "6  ('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...   \n",
       "7                                         ('Login',)   \n",
       "8  ('StartLoanApply', 'ViewLoanApplyIntro', 'Open...   \n",
       "9  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "\n",
       "                                           processed  \n",
       "0  login, start loan apply, view loan apply intro...  \n",
       "1  open app, login, start loan apply, view loan a...  \n",
       "2  open app, login, start loan apply, view loan a...  \n",
       "3  open app, login, use loan manage, get credit i...  \n",
       "4  get credit information, use loan manage, get c...  \n",
       "5  login, start loan apply, view loan apply intro...  \n",
       "6  end loan apply, view loan apply intro, complet...  \n",
       "7                                              login  \n",
       "8  start loan apply, view loan apply intro, open ...  \n",
       "9  open app, login, start loan apply, view loan a...  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## 전처리 0\n",
    "event_dict = {\n",
    "    'SignUp': 'sign up',\n",
    "    'OpenApp': 'open app',\n",
    "    'Login': 'login',\n",
    "    'ViewLoanApplyIntro': 'view loan apply intro',\n",
    "    'StartLoanApply': 'start loan apply',\n",
    "    'CompleteIDCertification': 'complete id certification',\n",
    "    'EndLoanApply': 'end loan apply',\n",
    "    'UseLoanManage': 'use loan manage',\n",
    "    'UsePrepayCalc': 'use prepay calculator',\n",
    "    'UseDSRCalc': 'use debt service ratio calculator',\n",
    "    'GetCreditInfo': 'get credit information'\n",
    "}\n",
    "\n",
    "## 전처리 1\n",
    "unique_paths['processed'] = unique_paths['path'].map(lambda x: x[1:-1])\n",
    "unique_paths['processed'] = unique_paths['processed'].map(lambda x: [item[1:-1].replace(\"'\", \"\") for item in x.split(', ')])\n",
    "\n",
    "## 전처리 2\n",
    "unique_paths['processed'] = unique_paths['processed'].map(lambda x: ', '.join([event_dict[item] for item in x]))\n",
    "unique_paths.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "UeHv9E1od31L",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UeHv9E1od31L",
    "outputId": "b0335b9a-392c-41e8-9201-42dba7d0c27b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 277577/277577 [01:31<00:00, 3028.73it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "277577\n",
      "TaggedDocument<['login', ',', 'start', 'loan', 'apply', ',', 'view', 'loan', 'apply', 'intro', ',', 'complete', 'id', 'certification', ',', 'end', 'loan', 'apply'], ['0']>\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "idx = 0\n",
    "tagged_data = []\n",
    "for path in tqdm(list(unique_paths['processed'])):\n",
    "    tagged_data.append(TaggedDocument(words=word_tokenize(path.lower()), tags=[str(idx)]))\n",
    "    idx += 1\n",
    "\n",
    "print(len(tagged_data))\n",
    "print(tagged_data[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fazUs1Agd7eQ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fazUs1Agd7eQ",
    "outputId": "a25c6d75-50c3-45d8-916f-9bf1de36e263"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  7%|▋         | 1/15 [05:30<1:17:08, 330.61s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 13%|█▎        | 2/15 [10:56<1:11:03, 328.00s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 20%|██        | 3/15 [16:28<1:05:56, 329.72s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 27%|██▋       | 4/15 [21:59<1:00:30, 330.05s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 33%|███▎      | 5/15 [27:23<54:39, 327.93s/it]  WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 40%|████      | 6/15 [32:49<49:07, 327.49s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 47%|████▋     | 7/15 [38:18<43:42, 327.78s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 53%|█████▎    | 8/15 [43:45<38:13, 327.71s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 60%|██████    | 9/15 [49:14<32:47, 327.99s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 67%|██████▋   | 10/15 [54:42<27:20, 328.02s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 73%|███████▎  | 11/15 [1:00:09<21:50, 327.55s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 80%|████████  | 12/15 [1:05:37<16:23, 327.92s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 87%|████████▋ | 13/15 [1:11:04<10:55, 327.66s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      " 93%|█████████▎| 14/15 [1:16:35<05:28, 328.66s/it]WARNING:gensim.models.word2vec:Effective 'alpha' higher than previous training cycles\n",
      "100%|██████████| 15/15 [1:22:03<00:00, 328.22s/it]\n"
     ]
    }
   ],
   "source": [
    "max_epochs = 15\n",
    "vector_size = 32\n",
    "model = Doc2Vec(vector_size=vector_size)\n",
    "model.build_vocab(tagged_data)\n",
    "\n",
    "for epoch in tqdm(range(max_epochs)):\n",
    "    model.train(tagged_data, total_examples=model.corpus_count, epochs=model.epochs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "wzKgk2KseACZ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wzKgk2KseACZ",
    "outputId": "e4d00519-aea8-438b-e83d-4f0dbbd62f00"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Saved\n",
      "Model Loaded\n"
     ]
    }
   ],
   "source": [
    "model.save(\"../Model/d2v.model\")\n",
    "print(\"Model Saved\")\n",
    "model = Doc2Vec.load(\"../Model/d2v.model\")\n",
    "print(\"Model Loaded\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1c41cac",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:44:03.500549Z",
     "iopub.status.busy": "2022-10-06T10:44:03.500214Z",
     "iopub.status.idle": "2022-10-06T10:47:12.519038Z",
     "shell.execute_reply": "2022-10-06T10:47:12.517659Z"
    },
    "id": "b1c41cac",
    "outputId": "3d871121-55cf-4aea-8fc5-e995736d8bd7",
    "papermill": {
     "duration": 189.028152,
     "end_time": "2022-10-06T10:47:12.521739",
     "exception": false,
     "start_time": "2022-10-06T10:44:03.493587",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 277577/277577 [04:35<00:00, 1008.23it/s]\n"
     ]
    }
   ],
   "source": [
    "path_embed_dict = dict()\n",
    "for path in tqdm(unique_paths['processed'].unique()):\n",
    "    path_embed_dict[path] = model.infer_vector(word_tokenize(path.lower()))\n",
    "unique_paths['embeddings'] = unique_paths['processed'].map(path_embed_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5838d73d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:47:12.731832Z",
     "iopub.status.busy": "2022-10-06T10:47:12.731462Z",
     "iopub.status.idle": "2022-10-06T10:47:12.736249Z",
     "shell.execute_reply": "2022-10-06T10:47:12.735366Z"
    },
    "id": "5838d73d",
    "outputId": "2b258d02-7ef1-43ad-d2ec-4e1866d636ad",
    "papermill": {
     "duration": 0.11142,
     "end_time": "2022-10-06T10:47:12.738076",
     "exception": false,
     "start_time": "2022-10-06T10:47:12.626656",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32\n"
     ]
    }
   ],
   "source": [
    "vector_size = unique_paths['embeddings'].iloc[0].shape[0]\n",
    "print(vector_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "20d911fc",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:47:12.939681Z",
     "iopub.status.busy": "2022-10-06T10:47:12.939314Z",
     "iopub.status.idle": "2022-10-06T10:55:41.392252Z",
     "shell.execute_reply": "2022-10-06T10:55:41.391354Z"
    },
    "id": "20d911fc",
    "outputId": "3a596b70-3d24-4611-9e85-06fed7298c96",
    "papermill": {
     "duration": 508.659849,
     "end_time": "2022-10-06T10:55:41.498297",
     "exception": false,
     "start_time": "2022-10-06T10:47:12.838448",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 32/32 [01:25<00:00,  2.67s/it]\n"
     ]
    }
   ],
   "source": [
    "components = []\n",
    "for i in tqdm(range(vector_size)):\n",
    "    item = [unique_paths['embeddings'].iloc[j][i] for j in range(unique_paths.shape[0])]\n",
    "    components.append(item)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "db043b5d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:55:41.710785Z",
     "iopub.status.busy": "2022-10-06T10:55:41.710185Z",
     "iopub.status.idle": "2022-10-06T10:55:51.300982Z",
     "shell.execute_reply": "2022-10-06T10:55:51.299905Z"
    },
    "id": "db043b5d",
    "outputId": "5cd359ca-7433-4ab2-c21f-eb9271486cd8",
    "papermill": {
     "duration": 9.708545,
     "end_time": "2022-10-06T10:55:51.311193",
     "exception": false,
     "start_time": "2022-10-06T10:55:41.602648",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "32it [00:02, 15.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(277577, 35)\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 277577 entries, 0 to 277576\n",
      "Data columns (total 35 columns):\n",
      " #   Column       Non-Null Count   Dtype  \n",
      "---  ------       --------------   -----  \n",
      " 0   path         277577 non-null  object \n",
      " 1   processed    277577 non-null  object \n",
      " 2   embeddings   277577 non-null  object \n",
      " 3   Component1   277577 non-null  float32\n",
      " 4   Component2   277577 non-null  float32\n",
      " 5   Component3   277577 non-null  float32\n",
      " 6   Component4   277577 non-null  float32\n",
      " 7   Component5   277577 non-null  float32\n",
      " 8   Component6   277577 non-null  float32\n",
      " 9   Component7   277577 non-null  float32\n",
      " 10  Component8   277577 non-null  float32\n",
      " 11  Component9   277577 non-null  float32\n",
      " 12  Component10  277577 non-null  float32\n",
      " 13  Component11  277577 non-null  float32\n",
      " 14  Component12  277577 non-null  float32\n",
      " 15  Component13  277577 non-null  float32\n",
      " 16  Component14  277577 non-null  float32\n",
      " 17  Component15  277577 non-null  float32\n",
      " 18  Component16  277577 non-null  float32\n",
      " 19  Component17  277577 non-null  float32\n",
      " 20  Component18  277577 non-null  float32\n",
      " 21  Component19  277577 non-null  float32\n",
      " 22  Component20  277577 non-null  float32\n",
      " 23  Component21  277577 non-null  float32\n",
      " 24  Component22  277577 non-null  float32\n",
      " 25  Component23  277577 non-null  float32\n",
      " 26  Component24  277577 non-null  float32\n",
      " 27  Component25  277577 non-null  float32\n",
      " 28  Component26  277577 non-null  float32\n",
      " 29  Component27  277577 non-null  float32\n",
      " 30  Component28  277577 non-null  float32\n",
      " 31  Component29  277577 non-null  float32\n",
      " 32  Component30  277577 non-null  float32\n",
      " 33  Component31  277577 non-null  float32\n",
      " 34  Component32  277577 non-null  float32\n",
      "dtypes: float32(32), object(3)\n",
      "memory usage: 40.2+ MB\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-8b75568e-b462-462e-b068-1c4648c769a9\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>path</th>\n",
       "      <th>processed</th>\n",
       "      <th>embeddings</th>\n",
       "      <th>Component1</th>\n",
       "      <th>Component2</th>\n",
       "      <th>Component3</th>\n",
       "      <th>Component4</th>\n",
       "      <th>Component5</th>\n",
       "      <th>Component6</th>\n",
       "      <th>Component7</th>\n",
       "      <th>Component8</th>\n",
       "      <th>Component9</th>\n",
       "      <th>Component10</th>\n",
       "      <th>Component11</th>\n",
       "      <th>Component12</th>\n",
       "      <th>Component13</th>\n",
       "      <th>Component14</th>\n",
       "      <th>Component15</th>\n",
       "      <th>Component16</th>\n",
       "      <th>Component17</th>\n",
       "      <th>Component18</th>\n",
       "      <th>Component19</th>\n",
       "      <th>Component20</th>\n",
       "      <th>Component21</th>\n",
       "      <th>Component22</th>\n",
       "      <th>Component23</th>\n",
       "      <th>Component24</th>\n",
       "      <th>Component25</th>\n",
       "      <th>Component26</th>\n",
       "      <th>Component27</th>\n",
       "      <th>Component28</th>\n",
       "      <th>Component29</th>\n",
       "      <th>Component30</th>\n",
       "      <th>Component31</th>\n",
       "      <th>Component32</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "      <td>[0.06371364, -0.0027174007, -0.0043107932, 0.0...</td>\n",
       "      <td>0.063714</td>\n",
       "      <td>-0.002717</td>\n",
       "      <td>-0.004311</td>\n",
       "      <td>0.012863</td>\n",
       "      <td>0.078528</td>\n",
       "      <td>0.048433</td>\n",
       "      <td>0.002448</td>\n",
       "      <td>0.059343</td>\n",
       "      <td>-0.106630</td>\n",
       "      <td>-0.127839</td>\n",
       "      <td>-0.096116</td>\n",
       "      <td>0.042791</td>\n",
       "      <td>-0.055906</td>\n",
       "      <td>-0.107747</td>\n",
       "      <td>0.163747</td>\n",
       "      <td>-0.073270</td>\n",
       "      <td>0.097261</td>\n",
       "      <td>0.001194</td>\n",
       "      <td>0.183316</td>\n",
       "      <td>-0.056198</td>\n",
       "      <td>-0.026070</td>\n",
       "      <td>-0.010077</td>\n",
       "      <td>0.057767</td>\n",
       "      <td>-0.010817</td>\n",
       "      <td>0.034083</td>\n",
       "      <td>0.079074</td>\n",
       "      <td>-0.017282</td>\n",
       "      <td>0.132538</td>\n",
       "      <td>-0.119630</td>\n",
       "      <td>0.032354</td>\n",
       "      <td>-0.074276</td>\n",
       "      <td>-0.039071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[-0.006433299, -0.012349155, 0.042309713, -0.0...</td>\n",
       "      <td>-0.006433</td>\n",
       "      <td>-0.012349</td>\n",
       "      <td>0.042310</td>\n",
       "      <td>-0.074814</td>\n",
       "      <td>0.044032</td>\n",
       "      <td>-0.065241</td>\n",
       "      <td>-0.118317</td>\n",
       "      <td>-0.009805</td>\n",
       "      <td>0.055622</td>\n",
       "      <td>-0.178393</td>\n",
       "      <td>-0.182844</td>\n",
       "      <td>0.043743</td>\n",
       "      <td>0.108397</td>\n",
       "      <td>-0.119072</td>\n",
       "      <td>0.298776</td>\n",
       "      <td>-0.120180</td>\n",
       "      <td>-0.013808</td>\n",
       "      <td>-0.053277</td>\n",
       "      <td>0.283966</td>\n",
       "      <td>-0.279235</td>\n",
       "      <td>0.065307</td>\n",
       "      <td>-0.146740</td>\n",
       "      <td>0.032129</td>\n",
       "      <td>-0.095018</td>\n",
       "      <td>-0.030789</td>\n",
       "      <td>0.191273</td>\n",
       "      <td>-0.109182</td>\n",
       "      <td>0.206062</td>\n",
       "      <td>-0.282896</td>\n",
       "      <td>0.099974</td>\n",
       "      <td>0.008139</td>\n",
       "      <td>-0.049811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[0.054358125, 0.05631258, 0.024874503, -0.1075...</td>\n",
       "      <td>0.054358</td>\n",
       "      <td>0.056313</td>\n",
       "      <td>0.024875</td>\n",
       "      <td>-0.107599</td>\n",
       "      <td>0.041433</td>\n",
       "      <td>0.083705</td>\n",
       "      <td>-0.108054</td>\n",
       "      <td>-0.002760</td>\n",
       "      <td>0.043527</td>\n",
       "      <td>-0.046906</td>\n",
       "      <td>-0.040594</td>\n",
       "      <td>-0.091720</td>\n",
       "      <td>0.016219</td>\n",
       "      <td>0.024631</td>\n",
       "      <td>0.102074</td>\n",
       "      <td>-0.005710</td>\n",
       "      <td>-0.029763</td>\n",
       "      <td>-0.192161</td>\n",
       "      <td>0.161236</td>\n",
       "      <td>-0.206023</td>\n",
       "      <td>0.042199</td>\n",
       "      <td>-0.006985</td>\n",
       "      <td>0.001805</td>\n",
       "      <td>0.109591</td>\n",
       "      <td>0.070431</td>\n",
       "      <td>0.127238</td>\n",
       "      <td>0.021463</td>\n",
       "      <td>0.101687</td>\n",
       "      <td>-0.088881</td>\n",
       "      <td>0.006124</td>\n",
       "      <td>-0.071933</td>\n",
       "      <td>-0.029166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>('OpenApp', 'Login', 'UseLoanManage', 'GetCred...</td>\n",
       "      <td>open app, login, use loan manage, get credit i...</td>\n",
       "      <td>[0.031055167, 0.071985036, -0.03701398, -0.053...</td>\n",
       "      <td>0.031055</td>\n",
       "      <td>0.071985</td>\n",
       "      <td>-0.037014</td>\n",
       "      <td>-0.053014</td>\n",
       "      <td>0.026453</td>\n",
       "      <td>0.051383</td>\n",
       "      <td>-0.043500</td>\n",
       "      <td>0.150232</td>\n",
       "      <td>-0.094052</td>\n",
       "      <td>-0.091896</td>\n",
       "      <td>-0.067998</td>\n",
       "      <td>-0.044945</td>\n",
       "      <td>-0.012982</td>\n",
       "      <td>-0.010849</td>\n",
       "      <td>0.072978</td>\n",
       "      <td>-0.041215</td>\n",
       "      <td>0.052298</td>\n",
       "      <td>-0.030508</td>\n",
       "      <td>0.154652</td>\n",
       "      <td>-0.102043</td>\n",
       "      <td>-0.058035</td>\n",
       "      <td>0.039161</td>\n",
       "      <td>0.061158</td>\n",
       "      <td>-0.029621</td>\n",
       "      <td>0.043033</td>\n",
       "      <td>0.075030</td>\n",
       "      <td>0.043423</td>\n",
       "      <td>0.091814</td>\n",
       "      <td>-0.109135</td>\n",
       "      <td>0.073924</td>\n",
       "      <td>-0.042635</td>\n",
       "      <td>-0.049726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>('GetCreditInfo', 'UseLoanManage', 'GetCreditI...</td>\n",
       "      <td>get credit information, use loan manage, get c...</td>\n",
       "      <td>[0.050136328, 0.07081204, -0.0013531668, -0.07...</td>\n",
       "      <td>0.050136</td>\n",
       "      <td>0.070812</td>\n",
       "      <td>-0.001353</td>\n",
       "      <td>-0.073679</td>\n",
       "      <td>-0.031094</td>\n",
       "      <td>0.127502</td>\n",
       "      <td>-0.002303</td>\n",
       "      <td>0.097203</td>\n",
       "      <td>-0.041101</td>\n",
       "      <td>-0.062483</td>\n",
       "      <td>-0.059103</td>\n",
       "      <td>-0.107579</td>\n",
       "      <td>-0.044388</td>\n",
       "      <td>0.004592</td>\n",
       "      <td>-0.010163</td>\n",
       "      <td>-0.051874</td>\n",
       "      <td>0.005546</td>\n",
       "      <td>-0.097045</td>\n",
       "      <td>0.125011</td>\n",
       "      <td>-0.062713</td>\n",
       "      <td>-0.113314</td>\n",
       "      <td>0.129667</td>\n",
       "      <td>0.014687</td>\n",
       "      <td>0.111240</td>\n",
       "      <td>0.009607</td>\n",
       "      <td>0.124755</td>\n",
       "      <td>0.129615</td>\n",
       "      <td>0.046244</td>\n",
       "      <td>-0.049326</td>\n",
       "      <td>0.012907</td>\n",
       "      <td>-0.077644</td>\n",
       "      <td>-0.079021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "      <td>[0.10427666, 0.011031533, -0.032765664, -0.050...</td>\n",
       "      <td>0.104277</td>\n",
       "      <td>0.011032</td>\n",
       "      <td>-0.032766</td>\n",
       "      <td>-0.050678</td>\n",
       "      <td>0.009463</td>\n",
       "      <td>0.047865</td>\n",
       "      <td>0.023726</td>\n",
       "      <td>-0.084342</td>\n",
       "      <td>0.012554</td>\n",
       "      <td>-0.168494</td>\n",
       "      <td>-0.097386</td>\n",
       "      <td>-0.081487</td>\n",
       "      <td>0.092587</td>\n",
       "      <td>-0.126993</td>\n",
       "      <td>0.128796</td>\n",
       "      <td>0.014674</td>\n",
       "      <td>0.144005</td>\n",
       "      <td>-0.100702</td>\n",
       "      <td>0.104164</td>\n",
       "      <td>-0.003518</td>\n",
       "      <td>0.016678</td>\n",
       "      <td>-0.114431</td>\n",
       "      <td>-0.102848</td>\n",
       "      <td>0.137504</td>\n",
       "      <td>0.035710</td>\n",
       "      <td>0.023495</td>\n",
       "      <td>0.054907</td>\n",
       "      <td>0.000656</td>\n",
       "      <td>-0.063511</td>\n",
       "      <td>-0.092066</td>\n",
       "      <td>-0.103760</td>\n",
       "      <td>-0.056617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...</td>\n",
       "      <td>end loan apply, view loan apply intro, complet...</td>\n",
       "      <td>[0.1052505, 0.06763772, 0.0043461267, 0.013259...</td>\n",
       "      <td>0.105251</td>\n",
       "      <td>0.067638</td>\n",
       "      <td>0.004346</td>\n",
       "      <td>0.013259</td>\n",
       "      <td>0.049477</td>\n",
       "      <td>-0.044369</td>\n",
       "      <td>0.059421</td>\n",
       "      <td>-0.017267</td>\n",
       "      <td>-0.067476</td>\n",
       "      <td>-0.153824</td>\n",
       "      <td>-0.142285</td>\n",
       "      <td>0.099692</td>\n",
       "      <td>-0.024039</td>\n",
       "      <td>-0.157745</td>\n",
       "      <td>0.087492</td>\n",
       "      <td>-0.010780</td>\n",
       "      <td>0.217738</td>\n",
       "      <td>0.010204</td>\n",
       "      <td>0.195356</td>\n",
       "      <td>-0.110219</td>\n",
       "      <td>-0.039211</td>\n",
       "      <td>-0.151496</td>\n",
       "      <td>0.042243</td>\n",
       "      <td>-0.061639</td>\n",
       "      <td>0.037335</td>\n",
       "      <td>0.105429</td>\n",
       "      <td>-0.062406</td>\n",
       "      <td>0.053244</td>\n",
       "      <td>-0.139738</td>\n",
       "      <td>0.043533</td>\n",
       "      <td>-0.021184</td>\n",
       "      <td>0.059896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>('Login',)</td>\n",
       "      <td>login</td>\n",
       "      <td>[0.053232253, 0.07727824, -0.040143024, -0.052...</td>\n",
       "      <td>0.053232</td>\n",
       "      <td>0.077278</td>\n",
       "      <td>-0.040143</td>\n",
       "      <td>-0.052889</td>\n",
       "      <td>0.016104</td>\n",
       "      <td>0.018469</td>\n",
       "      <td>-0.052565</td>\n",
       "      <td>0.046599</td>\n",
       "      <td>-0.067784</td>\n",
       "      <td>-0.112384</td>\n",
       "      <td>-0.070892</td>\n",
       "      <td>-0.022191</td>\n",
       "      <td>-0.009099</td>\n",
       "      <td>0.002649</td>\n",
       "      <td>0.072151</td>\n",
       "      <td>0.024153</td>\n",
       "      <td>0.100830</td>\n",
       "      <td>-0.016991</td>\n",
       "      <td>0.137095</td>\n",
       "      <td>-0.060045</td>\n",
       "      <td>-0.055566</td>\n",
       "      <td>-0.062975</td>\n",
       "      <td>0.065279</td>\n",
       "      <td>0.012828</td>\n",
       "      <td>0.007876</td>\n",
       "      <td>0.055051</td>\n",
       "      <td>-0.023394</td>\n",
       "      <td>0.081747</td>\n",
       "      <td>-0.114850</td>\n",
       "      <td>0.027080</td>\n",
       "      <td>-0.057619</td>\n",
       "      <td>-0.033613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>('StartLoanApply', 'ViewLoanApplyIntro', 'Open...</td>\n",
       "      <td>start loan apply, view loan apply intro, open ...</td>\n",
       "      <td>[-0.0016708188, 0.10211807, 0.018878523, -0.01...</td>\n",
       "      <td>-0.001671</td>\n",
       "      <td>0.102118</td>\n",
       "      <td>0.018879</td>\n",
       "      <td>-0.014849</td>\n",
       "      <td>0.047591</td>\n",
       "      <td>-0.077835</td>\n",
       "      <td>0.007995</td>\n",
       "      <td>0.068928</td>\n",
       "      <td>-0.043145</td>\n",
       "      <td>-0.151496</td>\n",
       "      <td>-0.129461</td>\n",
       "      <td>-0.063685</td>\n",
       "      <td>0.084314</td>\n",
       "      <td>-0.146427</td>\n",
       "      <td>0.097307</td>\n",
       "      <td>-0.110862</td>\n",
       "      <td>0.070262</td>\n",
       "      <td>-0.228543</td>\n",
       "      <td>0.253672</td>\n",
       "      <td>-0.226056</td>\n",
       "      <td>0.094342</td>\n",
       "      <td>-0.148660</td>\n",
       "      <td>0.043638</td>\n",
       "      <td>-0.059827</td>\n",
       "      <td>0.138846</td>\n",
       "      <td>0.031970</td>\n",
       "      <td>-0.100223</td>\n",
       "      <td>0.077079</td>\n",
       "      <td>-0.133173</td>\n",
       "      <td>0.042750</td>\n",
       "      <td>0.007902</td>\n",
       "      <td>-0.004513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[0.04850869, 0.13641827, -0.0004950871, 0.0082...</td>\n",
       "      <td>0.048509</td>\n",
       "      <td>0.136418</td>\n",
       "      <td>-0.000495</td>\n",
       "      <td>0.008273</td>\n",
       "      <td>0.084928</td>\n",
       "      <td>-0.031283</td>\n",
       "      <td>-0.036615</td>\n",
       "      <td>0.175278</td>\n",
       "      <td>-0.085980</td>\n",
       "      <td>-0.147543</td>\n",
       "      <td>-0.099325</td>\n",
       "      <td>0.055207</td>\n",
       "      <td>-0.115674</td>\n",
       "      <td>-0.055328</td>\n",
       "      <td>0.014914</td>\n",
       "      <td>-0.039102</td>\n",
       "      <td>0.144376</td>\n",
       "      <td>-0.045969</td>\n",
       "      <td>0.288825</td>\n",
       "      <td>-0.270984</td>\n",
       "      <td>-0.107616</td>\n",
       "      <td>-0.108731</td>\n",
       "      <td>0.175881</td>\n",
       "      <td>-0.109261</td>\n",
       "      <td>0.068311</td>\n",
       "      <td>0.061233</td>\n",
       "      <td>-0.067302</td>\n",
       "      <td>0.151313</td>\n",
       "      <td>-0.097409</td>\n",
       "      <td>0.130504</td>\n",
       "      <td>0.031902</td>\n",
       "      <td>0.043764</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8b75568e-b462-462e-b068-1c4648c769a9')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-8b75568e-b462-462e-b068-1c4648c769a9 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-8b75568e-b462-462e-b068-1c4648c769a9');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "                                                path  \\\n",
       "0  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "1  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "2  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "3  ('OpenApp', 'Login', 'UseLoanManage', 'GetCred...   \n",
       "4  ('GetCreditInfo', 'UseLoanManage', 'GetCreditI...   \n",
       "5  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "6  ('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...   \n",
       "7                                         ('Login',)   \n",
       "8  ('StartLoanApply', 'ViewLoanApplyIntro', 'Open...   \n",
       "9  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "\n",
       "                                           processed  \\\n",
       "0  login, start loan apply, view loan apply intro...   \n",
       "1  open app, login, start loan apply, view loan a...   \n",
       "2  open app, login, start loan apply, view loan a...   \n",
       "3  open app, login, use loan manage, get credit i...   \n",
       "4  get credit information, use loan manage, get c...   \n",
       "5  login, start loan apply, view loan apply intro...   \n",
       "6  end loan apply, view loan apply intro, complet...   \n",
       "7                                              login   \n",
       "8  start loan apply, view loan apply intro, open ...   \n",
       "9  open app, login, start loan apply, view loan a...   \n",
       "\n",
       "                                          embeddings  Component1  Component2  \\\n",
       "0  [0.06371364, -0.0027174007, -0.0043107932, 0.0...    0.063714   -0.002717   \n",
       "1  [-0.006433299, -0.012349155, 0.042309713, -0.0...   -0.006433   -0.012349   \n",
       "2  [0.054358125, 0.05631258, 0.024874503, -0.1075...    0.054358    0.056313   \n",
       "3  [0.031055167, 0.071985036, -0.03701398, -0.053...    0.031055    0.071985   \n",
       "4  [0.050136328, 0.07081204, -0.0013531668, -0.07...    0.050136    0.070812   \n",
       "5  [0.10427666, 0.011031533, -0.032765664, -0.050...    0.104277    0.011032   \n",
       "6  [0.1052505, 0.06763772, 0.0043461267, 0.013259...    0.105251    0.067638   \n",
       "7  [0.053232253, 0.07727824, -0.040143024, -0.052...    0.053232    0.077278   \n",
       "8  [-0.0016708188, 0.10211807, 0.018878523, -0.01...   -0.001671    0.102118   \n",
       "9  [0.04850869, 0.13641827, -0.0004950871, 0.0082...    0.048509    0.136418   \n",
       "\n",
       "   Component3  Component4  Component5  Component6  Component7  Component8  \\\n",
       "0   -0.004311    0.012863    0.078528    0.048433    0.002448    0.059343   \n",
       "1    0.042310   -0.074814    0.044032   -0.065241   -0.118317   -0.009805   \n",
       "2    0.024875   -0.107599    0.041433    0.083705   -0.108054   -0.002760   \n",
       "3   -0.037014   -0.053014    0.026453    0.051383   -0.043500    0.150232   \n",
       "4   -0.001353   -0.073679   -0.031094    0.127502   -0.002303    0.097203   \n",
       "5   -0.032766   -0.050678    0.009463    0.047865    0.023726   -0.084342   \n",
       "6    0.004346    0.013259    0.049477   -0.044369    0.059421   -0.017267   \n",
       "7   -0.040143   -0.052889    0.016104    0.018469   -0.052565    0.046599   \n",
       "8    0.018879   -0.014849    0.047591   -0.077835    0.007995    0.068928   \n",
       "9   -0.000495    0.008273    0.084928   -0.031283   -0.036615    0.175278   \n",
       "\n",
       "   Component9  Component10  Component11  Component12  Component13  \\\n",
       "0   -0.106630    -0.127839    -0.096116     0.042791    -0.055906   \n",
       "1    0.055622    -0.178393    -0.182844     0.043743     0.108397   \n",
       "2    0.043527    -0.046906    -0.040594    -0.091720     0.016219   \n",
       "3   -0.094052    -0.091896    -0.067998    -0.044945    -0.012982   \n",
       "4   -0.041101    -0.062483    -0.059103    -0.107579    -0.044388   \n",
       "5    0.012554    -0.168494    -0.097386    -0.081487     0.092587   \n",
       "6   -0.067476    -0.153824    -0.142285     0.099692    -0.024039   \n",
       "7   -0.067784    -0.112384    -0.070892    -0.022191    -0.009099   \n",
       "8   -0.043145    -0.151496    -0.129461    -0.063685     0.084314   \n",
       "9   -0.085980    -0.147543    -0.099325     0.055207    -0.115674   \n",
       "\n",
       "   Component14  Component15  Component16  Component17  Component18  \\\n",
       "0    -0.107747     0.163747    -0.073270     0.097261     0.001194   \n",
       "1    -0.119072     0.298776    -0.120180    -0.013808    -0.053277   \n",
       "2     0.024631     0.102074    -0.005710    -0.029763    -0.192161   \n",
       "3    -0.010849     0.072978    -0.041215     0.052298    -0.030508   \n",
       "4     0.004592    -0.010163    -0.051874     0.005546    -0.097045   \n",
       "5    -0.126993     0.128796     0.014674     0.144005    -0.100702   \n",
       "6    -0.157745     0.087492    -0.010780     0.217738     0.010204   \n",
       "7     0.002649     0.072151     0.024153     0.100830    -0.016991   \n",
       "8    -0.146427     0.097307    -0.110862     0.070262    -0.228543   \n",
       "9    -0.055328     0.014914    -0.039102     0.144376    -0.045969   \n",
       "\n",
       "   Component19  Component20  Component21  Component22  Component23  \\\n",
       "0     0.183316    -0.056198    -0.026070    -0.010077     0.057767   \n",
       "1     0.283966    -0.279235     0.065307    -0.146740     0.032129   \n",
       "2     0.161236    -0.206023     0.042199    -0.006985     0.001805   \n",
       "3     0.154652    -0.102043    -0.058035     0.039161     0.061158   \n",
       "4     0.125011    -0.062713    -0.113314     0.129667     0.014687   \n",
       "5     0.104164    -0.003518     0.016678    -0.114431    -0.102848   \n",
       "6     0.195356    -0.110219    -0.039211    -0.151496     0.042243   \n",
       "7     0.137095    -0.060045    -0.055566    -0.062975     0.065279   \n",
       "8     0.253672    -0.226056     0.094342    -0.148660     0.043638   \n",
       "9     0.288825    -0.270984    -0.107616    -0.108731     0.175881   \n",
       "\n",
       "   Component24  Component25  Component26  Component27  Component28  \\\n",
       "0    -0.010817     0.034083     0.079074    -0.017282     0.132538   \n",
       "1    -0.095018    -0.030789     0.191273    -0.109182     0.206062   \n",
       "2     0.109591     0.070431     0.127238     0.021463     0.101687   \n",
       "3    -0.029621     0.043033     0.075030     0.043423     0.091814   \n",
       "4     0.111240     0.009607     0.124755     0.129615     0.046244   \n",
       "5     0.137504     0.035710     0.023495     0.054907     0.000656   \n",
       "6    -0.061639     0.037335     0.105429    -0.062406     0.053244   \n",
       "7     0.012828     0.007876     0.055051    -0.023394     0.081747   \n",
       "8    -0.059827     0.138846     0.031970    -0.100223     0.077079   \n",
       "9    -0.109261     0.068311     0.061233    -0.067302     0.151313   \n",
       "\n",
       "   Component29  Component30  Component31  Component32  \n",
       "0    -0.119630     0.032354    -0.074276    -0.039071  \n",
       "1    -0.282896     0.099974     0.008139    -0.049811  \n",
       "2    -0.088881     0.006124    -0.071933    -0.029166  \n",
       "3    -0.109135     0.073924    -0.042635    -0.049726  \n",
       "4    -0.049326     0.012907    -0.077644    -0.079021  \n",
       "5    -0.063511    -0.092066    -0.103760    -0.056617  \n",
       "6    -0.139738     0.043533    -0.021184     0.059896  \n",
       "7    -0.114850     0.027080    -0.057619    -0.033613  \n",
       "8    -0.133173     0.042750     0.007902    -0.004513  \n",
       "9    -0.097409     0.130504     0.031902     0.043764  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i, item in tqdm(enumerate(components)):\n",
    "    unique_paths[f\"Component{i+1}\"] = item\n",
    "viewinfo(unique_paths)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2NVaZoYQ2xp9",
   "metadata": {
    "id": "2NVaZoYQ2xp9"
   },
   "outputs": [],
   "source": [
    "# for col in unique_paths.columns.values:\n",
    "#     if \"Component\" in col:\n",
    "#         unique_paths[col] = minmax(unique_paths[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cabf9d8d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T10:55:51.553491Z",
     "iopub.status.busy": "2022-10-06T10:55:51.553141Z",
     "iopub.status.idle": "2022-10-06T10:55:51.883246Z",
     "shell.execute_reply": "2022-10-06T10:55:51.882360Z"
    },
    "id": "cabf9d8d",
    "papermill": {
     "duration": 0.459405,
     "end_time": "2022-10-06T10:55:51.885327",
     "exception": false,
     "start_time": "2022-10-06T10:55:51.425922",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "keep_keys= ['path', 'processed', 'embeddings']\n",
    "to_keep = unique_paths[keep_keys]\n",
    "to_use = unique_paths.drop(keep_keys, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "408eccb9",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 595
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T10:55:52.108836Z",
     "iopub.status.busy": "2022-10-06T10:55:52.108479Z",
     "iopub.status.idle": "2022-10-06T10:55:52.130012Z",
     "shell.execute_reply": "2022-10-06T10:55:52.128666Z"
    },
    "id": "408eccb9",
    "outputId": "fb083b57-b735-4a29-9cd4-bbf5b847c49f",
    "papermill": {
     "duration": 0.13531,
     "end_time": "2022-10-06T10:55:52.131847",
     "exception": false,
     "start_time": "2022-10-06T10:55:51.996537",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(277577, 3)\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 277577 entries, 0 to 277576\n",
      "Data columns (total 3 columns):\n",
      " #   Column      Non-Null Count   Dtype \n",
      "---  ------      --------------   ----- \n",
      " 0   path        277577 non-null  object\n",
      " 1   processed   277577 non-null  object\n",
      " 2   embeddings  277577 non-null  object\n",
      "dtypes: object(3)\n",
      "memory usage: 6.4+ MB\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-7bee2e97-9720-48cf-8011-88eec461abfd\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>path</th>\n",
       "      <th>processed</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "      <td>[0.06371364, -0.0027174007, -0.0043107932, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[-0.006433299, -0.012349155, 0.042309713, -0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[0.054358125, 0.05631258, 0.024874503, -0.1075...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>('OpenApp', 'Login', 'UseLoanManage', 'GetCred...</td>\n",
       "      <td>open app, login, use loan manage, get credit i...</td>\n",
       "      <td>[0.031055167, 0.071985036, -0.03701398, -0.053...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>('GetCreditInfo', 'UseLoanManage', 'GetCreditI...</td>\n",
       "      <td>get credit information, use loan manage, get c...</td>\n",
       "      <td>[0.050136328, 0.07081204, -0.0013531668, -0.07...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>('Login', 'StartLoanApply', 'ViewLoanApplyIntr...</td>\n",
       "      <td>login, start loan apply, view loan apply intro...</td>\n",
       "      <td>[0.10427666, 0.011031533, -0.032765664, -0.050...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...</td>\n",
       "      <td>end loan apply, view loan apply intro, complet...</td>\n",
       "      <td>[0.1052505, 0.06763772, 0.0043461267, 0.013259...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>('Login',)</td>\n",
       "      <td>login</td>\n",
       "      <td>[0.053232253, 0.07727824, -0.040143024, -0.052...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>('StartLoanApply', 'ViewLoanApplyIntro', 'Open...</td>\n",
       "      <td>start loan apply, view loan apply intro, open ...</td>\n",
       "      <td>[-0.0016708188, 0.10211807, 0.018878523, -0.01...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...</td>\n",
       "      <td>open app, login, start loan apply, view loan a...</td>\n",
       "      <td>[0.04850869, 0.13641827, -0.0004950871, 0.0082...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7bee2e97-9720-48cf-8011-88eec461abfd')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-7bee2e97-9720-48cf-8011-88eec461abfd button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-7bee2e97-9720-48cf-8011-88eec461abfd');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "                                                path  \\\n",
       "0  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "1  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "2  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "3  ('OpenApp', 'Login', 'UseLoanManage', 'GetCred...   \n",
       "4  ('GetCreditInfo', 'UseLoanManage', 'GetCreditI...   \n",
       "5  ('Login', 'StartLoanApply', 'ViewLoanApplyIntr...   \n",
       "6  ('EndLoanApply', 'ViewLoanApplyIntro', 'Comple...   \n",
       "7                                         ('Login',)   \n",
       "8  ('StartLoanApply', 'ViewLoanApplyIntro', 'Open...   \n",
       "9  ('OpenApp', 'Login', 'StartLoanApply', 'ViewLo...   \n",
       "\n",
       "                                           processed  \\\n",
       "0  login, start loan apply, view loan apply intro...   \n",
       "1  open app, login, start loan apply, view loan a...   \n",
       "2  open app, login, start loan apply, view loan a...   \n",
       "3  open app, login, use loan manage, get credit i...   \n",
       "4  get credit information, use loan manage, get c...   \n",
       "5  login, start loan apply, view loan apply intro...   \n",
       "6  end loan apply, view loan apply intro, complet...   \n",
       "7                                              login   \n",
       "8  start loan apply, view loan apply intro, open ...   \n",
       "9  open app, login, start loan apply, view loan a...   \n",
       "\n",
       "                                          embeddings  \n",
       "0  [0.06371364, -0.0027174007, -0.0043107932, 0.0...  \n",
       "1  [-0.006433299, -0.012349155, 0.042309713, -0.0...  \n",
       "2  [0.054358125, 0.05631258, 0.024874503, -0.1075...  \n",
       "3  [0.031055167, 0.071985036, -0.03701398, -0.053...  \n",
       "4  [0.050136328, 0.07081204, -0.0013531668, -0.07...  \n",
       "5  [0.10427666, 0.011031533, -0.032765664, -0.050...  \n",
       "6  [0.1052505, 0.06763772, 0.0043461267, 0.013259...  \n",
       "7  [0.053232253, 0.07727824, -0.040143024, -0.052...  \n",
       "8  [-0.0016708188, 0.10211807, 0.018878523, -0.01...  \n",
       "9  [0.04850869, 0.13641827, -0.0004950871, 0.0082...  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "viewinfo(to_keep)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "rRTTC0Qk3Dmz",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "rRTTC0Qk3Dmz",
    "outputId": "1a1de2fe-b882-432d-94d9-b8fadee66169"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(277577, 32)\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 277577 entries, 0 to 277576\n",
      "Data columns (total 32 columns):\n",
      " #   Column       Non-Null Count   Dtype  \n",
      "---  ------       --------------   -----  \n",
      " 0   Component1   277577 non-null  float32\n",
      " 1   Component2   277577 non-null  float32\n",
      " 2   Component3   277577 non-null  float32\n",
      " 3   Component4   277577 non-null  float32\n",
      " 4   Component5   277577 non-null  float32\n",
      " 5   Component6   277577 non-null  float32\n",
      " 6   Component7   277577 non-null  float32\n",
      " 7   Component8   277577 non-null  float32\n",
      " 8   Component9   277577 non-null  float32\n",
      " 9   Component10  277577 non-null  float32\n",
      " 10  Component11  277577 non-null  float32\n",
      " 11  Component12  277577 non-null  float32\n",
      " 12  Component13  277577 non-null  float32\n",
      " 13  Component14  277577 non-null  float32\n",
      " 14  Component15  277577 non-null  float32\n",
      " 15  Component16  277577 non-null  float32\n",
      " 16  Component17  277577 non-null  float32\n",
      " 17  Component18  277577 non-null  float32\n",
      " 18  Component19  277577 non-null  float32\n",
      " 19  Component20  277577 non-null  float32\n",
      " 20  Component21  277577 non-null  float32\n",
      " 21  Component22  277577 non-null  float32\n",
      " 22  Component23  277577 non-null  float32\n",
      " 23  Component24  277577 non-null  float32\n",
      " 24  Component25  277577 non-null  float32\n",
      " 25  Component26  277577 non-null  float32\n",
      " 26  Component27  277577 non-null  float32\n",
      " 27  Component28  277577 non-null  float32\n",
      " 28  Component29  277577 non-null  float32\n",
      " 29  Component30  277577 non-null  float32\n",
      " 30  Component31  277577 non-null  float32\n",
      " 31  Component32  277577 non-null  float32\n",
      "dtypes: float32(32)\n",
      "memory usage: 33.9 MB\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-6eb4cc2b-a315-42b5-8bdd-8d2e840ba501\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Component1</th>\n",
       "      <th>Component2</th>\n",
       "      <th>Component3</th>\n",
       "      <th>Component4</th>\n",
       "      <th>Component5</th>\n",
       "      <th>Component6</th>\n",
       "      <th>Component7</th>\n",
       "      <th>Component8</th>\n",
       "      <th>Component9</th>\n",
       "      <th>Component10</th>\n",
       "      <th>Component11</th>\n",
       "      <th>Component12</th>\n",
       "      <th>Component13</th>\n",
       "      <th>Component14</th>\n",
       "      <th>Component15</th>\n",
       "      <th>Component16</th>\n",
       "      <th>Component17</th>\n",
       "      <th>Component18</th>\n",
       "      <th>Component19</th>\n",
       "      <th>Component20</th>\n",
       "      <th>Component21</th>\n",
       "      <th>Component22</th>\n",
       "      <th>Component23</th>\n",
       "      <th>Component24</th>\n",
       "      <th>Component25</th>\n",
       "      <th>Component26</th>\n",
       "      <th>Component27</th>\n",
       "      <th>Component28</th>\n",
       "      <th>Component29</th>\n",
       "      <th>Component30</th>\n",
       "      <th>Component31</th>\n",
       "      <th>Component32</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.063714</td>\n",
       "      <td>-0.002717</td>\n",
       "      <td>-0.004311</td>\n",
       "      <td>0.012863</td>\n",
       "      <td>0.078528</td>\n",
       "      <td>0.048433</td>\n",
       "      <td>0.002448</td>\n",
       "      <td>0.059343</td>\n",
       "      <td>-0.106630</td>\n",
       "      <td>-0.127839</td>\n",
       "      <td>-0.096116</td>\n",
       "      <td>0.042791</td>\n",
       "      <td>-0.055906</td>\n",
       "      <td>-0.107747</td>\n",
       "      <td>0.163747</td>\n",
       "      <td>-0.073270</td>\n",
       "      <td>0.097261</td>\n",
       "      <td>0.001194</td>\n",
       "      <td>0.183316</td>\n",
       "      <td>-0.056198</td>\n",
       "      <td>-0.026070</td>\n",
       "      <td>-0.010077</td>\n",
       "      <td>0.057767</td>\n",
       "      <td>-0.010817</td>\n",
       "      <td>0.034083</td>\n",
       "      <td>0.079074</td>\n",
       "      <td>-0.017282</td>\n",
       "      <td>0.132538</td>\n",
       "      <td>-0.119630</td>\n",
       "      <td>0.032354</td>\n",
       "      <td>-0.074276</td>\n",
       "      <td>-0.039071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.006433</td>\n",
       "      <td>-0.012349</td>\n",
       "      <td>0.042310</td>\n",
       "      <td>-0.074814</td>\n",
       "      <td>0.044032</td>\n",
       "      <td>-0.065241</td>\n",
       "      <td>-0.118317</td>\n",
       "      <td>-0.009805</td>\n",
       "      <td>0.055622</td>\n",
       "      <td>-0.178393</td>\n",
       "      <td>-0.182844</td>\n",
       "      <td>0.043743</td>\n",
       "      <td>0.108397</td>\n",
       "      <td>-0.119072</td>\n",
       "      <td>0.298776</td>\n",
       "      <td>-0.120180</td>\n",
       "      <td>-0.013808</td>\n",
       "      <td>-0.053277</td>\n",
       "      <td>0.283966</td>\n",
       "      <td>-0.279235</td>\n",
       "      <td>0.065307</td>\n",
       "      <td>-0.146740</td>\n",
       "      <td>0.032129</td>\n",
       "      <td>-0.095018</td>\n",
       "      <td>-0.030789</td>\n",
       "      <td>0.191273</td>\n",
       "      <td>-0.109182</td>\n",
       "      <td>0.206062</td>\n",
       "      <td>-0.282896</td>\n",
       "      <td>0.099974</td>\n",
       "      <td>0.008139</td>\n",
       "      <td>-0.049811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.054358</td>\n",
       "      <td>0.056313</td>\n",
       "      <td>0.024875</td>\n",
       "      <td>-0.107599</td>\n",
       "      <td>0.041433</td>\n",
       "      <td>0.083705</td>\n",
       "      <td>-0.108054</td>\n",
       "      <td>-0.002760</td>\n",
       "      <td>0.043527</td>\n",
       "      <td>-0.046906</td>\n",
       "      <td>-0.040594</td>\n",
       "      <td>-0.091720</td>\n",
       "      <td>0.016219</td>\n",
       "      <td>0.024631</td>\n",
       "      <td>0.102074</td>\n",
       "      <td>-0.005710</td>\n",
       "      <td>-0.029763</td>\n",
       "      <td>-0.192161</td>\n",
       "      <td>0.161236</td>\n",
       "      <td>-0.206023</td>\n",
       "      <td>0.042199</td>\n",
       "      <td>-0.006985</td>\n",
       "      <td>0.001805</td>\n",
       "      <td>0.109591</td>\n",
       "      <td>0.070431</td>\n",
       "      <td>0.127238</td>\n",
       "      <td>0.021463</td>\n",
       "      <td>0.101687</td>\n",
       "      <td>-0.088881</td>\n",
       "      <td>0.006124</td>\n",
       "      <td>-0.071933</td>\n",
       "      <td>-0.029166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.031055</td>\n",
       "      <td>0.071985</td>\n",
       "      <td>-0.037014</td>\n",
       "      <td>-0.053014</td>\n",
       "      <td>0.026453</td>\n",
       "      <td>0.051383</td>\n",
       "      <td>-0.043500</td>\n",
       "      <td>0.150232</td>\n",
       "      <td>-0.094052</td>\n",
       "      <td>-0.091896</td>\n",
       "      <td>-0.067998</td>\n",
       "      <td>-0.044945</td>\n",
       "      <td>-0.012982</td>\n",
       "      <td>-0.010849</td>\n",
       "      <td>0.072978</td>\n",
       "      <td>-0.041215</td>\n",
       "      <td>0.052298</td>\n",
       "      <td>-0.030508</td>\n",
       "      <td>0.154652</td>\n",
       "      <td>-0.102043</td>\n",
       "      <td>-0.058035</td>\n",
       "      <td>0.039161</td>\n",
       "      <td>0.061158</td>\n",
       "      <td>-0.029621</td>\n",
       "      <td>0.043033</td>\n",
       "      <td>0.075030</td>\n",
       "      <td>0.043423</td>\n",
       "      <td>0.091814</td>\n",
       "      <td>-0.109135</td>\n",
       "      <td>0.073924</td>\n",
       "      <td>-0.042635</td>\n",
       "      <td>-0.049726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.050136</td>\n",
       "      <td>0.070812</td>\n",
       "      <td>-0.001353</td>\n",
       "      <td>-0.073679</td>\n",
       "      <td>-0.031094</td>\n",
       "      <td>0.127502</td>\n",
       "      <td>-0.002303</td>\n",
       "      <td>0.097203</td>\n",
       "      <td>-0.041101</td>\n",
       "      <td>-0.062483</td>\n",
       "      <td>-0.059103</td>\n",
       "      <td>-0.107579</td>\n",
       "      <td>-0.044388</td>\n",
       "      <td>0.004592</td>\n",
       "      <td>-0.010163</td>\n",
       "      <td>-0.051874</td>\n",
       "      <td>0.005546</td>\n",
       "      <td>-0.097045</td>\n",
       "      <td>0.125011</td>\n",
       "      <td>-0.062713</td>\n",
       "      <td>-0.113314</td>\n",
       "      <td>0.129667</td>\n",
       "      <td>0.014687</td>\n",
       "      <td>0.111240</td>\n",
       "      <td>0.009607</td>\n",
       "      <td>0.124755</td>\n",
       "      <td>0.129615</td>\n",
       "      <td>0.046244</td>\n",
       "      <td>-0.049326</td>\n",
       "      <td>0.012907</td>\n",
       "      <td>-0.077644</td>\n",
       "      <td>-0.079021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.104277</td>\n",
       "      <td>0.011032</td>\n",
       "      <td>-0.032766</td>\n",
       "      <td>-0.050678</td>\n",
       "      <td>0.009463</td>\n",
       "      <td>0.047865</td>\n",
       "      <td>0.023726</td>\n",
       "      <td>-0.084342</td>\n",
       "      <td>0.012554</td>\n",
       "      <td>-0.168494</td>\n",
       "      <td>-0.097386</td>\n",
       "      <td>-0.081487</td>\n",
       "      <td>0.092587</td>\n",
       "      <td>-0.126993</td>\n",
       "      <td>0.128796</td>\n",
       "      <td>0.014674</td>\n",
       "      <td>0.144005</td>\n",
       "      <td>-0.100702</td>\n",
       "      <td>0.104164</td>\n",
       "      <td>-0.003518</td>\n",
       "      <td>0.016678</td>\n",
       "      <td>-0.114431</td>\n",
       "      <td>-0.102848</td>\n",
       "      <td>0.137504</td>\n",
       "      <td>0.035710</td>\n",
       "      <td>0.023495</td>\n",
       "      <td>0.054907</td>\n",
       "      <td>0.000656</td>\n",
       "      <td>-0.063511</td>\n",
       "      <td>-0.092066</td>\n",
       "      <td>-0.103760</td>\n",
       "      <td>-0.056617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.105251</td>\n",
       "      <td>0.067638</td>\n",
       "      <td>0.004346</td>\n",
       "      <td>0.013259</td>\n",
       "      <td>0.049477</td>\n",
       "      <td>-0.044369</td>\n",
       "      <td>0.059421</td>\n",
       "      <td>-0.017267</td>\n",
       "      <td>-0.067476</td>\n",
       "      <td>-0.153824</td>\n",
       "      <td>-0.142285</td>\n",
       "      <td>0.099692</td>\n",
       "      <td>-0.024039</td>\n",
       "      <td>-0.157745</td>\n",
       "      <td>0.087492</td>\n",
       "      <td>-0.010780</td>\n",
       "      <td>0.217738</td>\n",
       "      <td>0.010204</td>\n",
       "      <td>0.195356</td>\n",
       "      <td>-0.110219</td>\n",
       "      <td>-0.039211</td>\n",
       "      <td>-0.151496</td>\n",
       "      <td>0.042243</td>\n",
       "      <td>-0.061639</td>\n",
       "      <td>0.037335</td>\n",
       "      <td>0.105429</td>\n",
       "      <td>-0.062406</td>\n",
       "      <td>0.053244</td>\n",
       "      <td>-0.139738</td>\n",
       "      <td>0.043533</td>\n",
       "      <td>-0.021184</td>\n",
       "      <td>0.059896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.053232</td>\n",
       "      <td>0.077278</td>\n",
       "      <td>-0.040143</td>\n",
       "      <td>-0.052889</td>\n",
       "      <td>0.016104</td>\n",
       "      <td>0.018469</td>\n",
       "      <td>-0.052565</td>\n",
       "      <td>0.046599</td>\n",
       "      <td>-0.067784</td>\n",
       "      <td>-0.112384</td>\n",
       "      <td>-0.070892</td>\n",
       "      <td>-0.022191</td>\n",
       "      <td>-0.009099</td>\n",
       "      <td>0.002649</td>\n",
       "      <td>0.072151</td>\n",
       "      <td>0.024153</td>\n",
       "      <td>0.100830</td>\n",
       "      <td>-0.016991</td>\n",
       "      <td>0.137095</td>\n",
       "      <td>-0.060045</td>\n",
       "      <td>-0.055566</td>\n",
       "      <td>-0.062975</td>\n",
       "      <td>0.065279</td>\n",
       "      <td>0.012828</td>\n",
       "      <td>0.007876</td>\n",
       "      <td>0.055051</td>\n",
       "      <td>-0.023394</td>\n",
       "      <td>0.081747</td>\n",
       "      <td>-0.114850</td>\n",
       "      <td>0.027080</td>\n",
       "      <td>-0.057619</td>\n",
       "      <td>-0.033613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>-0.001671</td>\n",
       "      <td>0.102118</td>\n",
       "      <td>0.018879</td>\n",
       "      <td>-0.014849</td>\n",
       "      <td>0.047591</td>\n",
       "      <td>-0.077835</td>\n",
       "      <td>0.007995</td>\n",
       "      <td>0.068928</td>\n",
       "      <td>-0.043145</td>\n",
       "      <td>-0.151496</td>\n",
       "      <td>-0.129461</td>\n",
       "      <td>-0.063685</td>\n",
       "      <td>0.084314</td>\n",
       "      <td>-0.146427</td>\n",
       "      <td>0.097307</td>\n",
       "      <td>-0.110862</td>\n",
       "      <td>0.070262</td>\n",
       "      <td>-0.228543</td>\n",
       "      <td>0.253672</td>\n",
       "      <td>-0.226056</td>\n",
       "      <td>0.094342</td>\n",
       "      <td>-0.148660</td>\n",
       "      <td>0.043638</td>\n",
       "      <td>-0.059827</td>\n",
       "      <td>0.138846</td>\n",
       "      <td>0.031970</td>\n",
       "      <td>-0.100223</td>\n",
       "      <td>0.077079</td>\n",
       "      <td>-0.133173</td>\n",
       "      <td>0.042750</td>\n",
       "      <td>0.007902</td>\n",
       "      <td>-0.004513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.048509</td>\n",
       "      <td>0.136418</td>\n",
       "      <td>-0.000495</td>\n",
       "      <td>0.008273</td>\n",
       "      <td>0.084928</td>\n",
       "      <td>-0.031283</td>\n",
       "      <td>-0.036615</td>\n",
       "      <td>0.175278</td>\n",
       "      <td>-0.085980</td>\n",
       "      <td>-0.147543</td>\n",
       "      <td>-0.099325</td>\n",
       "      <td>0.055207</td>\n",
       "      <td>-0.115674</td>\n",
       "      <td>-0.055328</td>\n",
       "      <td>0.014914</td>\n",
       "      <td>-0.039102</td>\n",
       "      <td>0.144376</td>\n",
       "      <td>-0.045969</td>\n",
       "      <td>0.288825</td>\n",
       "      <td>-0.270984</td>\n",
       "      <td>-0.107616</td>\n",
       "      <td>-0.108731</td>\n",
       "      <td>0.175881</td>\n",
       "      <td>-0.109261</td>\n",
       "      <td>0.068311</td>\n",
       "      <td>0.061233</td>\n",
       "      <td>-0.067302</td>\n",
       "      <td>0.151313</td>\n",
       "      <td>-0.097409</td>\n",
       "      <td>0.130504</td>\n",
       "      <td>0.031902</td>\n",
       "      <td>0.043764</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6eb4cc2b-a315-42b5-8bdd-8d2e840ba501')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-6eb4cc2b-a315-42b5-8bdd-8d2e840ba501 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-6eb4cc2b-a315-42b5-8bdd-8d2e840ba501');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   Component1  Component2  Component3  Component4  Component5  Component6  \\\n",
       "0    0.063714   -0.002717   -0.004311    0.012863    0.078528    0.048433   \n",
       "1   -0.006433   -0.012349    0.042310   -0.074814    0.044032   -0.065241   \n",
       "2    0.054358    0.056313    0.024875   -0.107599    0.041433    0.083705   \n",
       "3    0.031055    0.071985   -0.037014   -0.053014    0.026453    0.051383   \n",
       "4    0.050136    0.070812   -0.001353   -0.073679   -0.031094    0.127502   \n",
       "5    0.104277    0.011032   -0.032766   -0.050678    0.009463    0.047865   \n",
       "6    0.105251    0.067638    0.004346    0.013259    0.049477   -0.044369   \n",
       "7    0.053232    0.077278   -0.040143   -0.052889    0.016104    0.018469   \n",
       "8   -0.001671    0.102118    0.018879   -0.014849    0.047591   -0.077835   \n",
       "9    0.048509    0.136418   -0.000495    0.008273    0.084928   -0.031283   \n",
       "\n",
       "   Component7  Component8  Component9  Component10  Component11  Component12  \\\n",
       "0    0.002448    0.059343   -0.106630    -0.127839    -0.096116     0.042791   \n",
       "1   -0.118317   -0.009805    0.055622    -0.178393    -0.182844     0.043743   \n",
       "2   -0.108054   -0.002760    0.043527    -0.046906    -0.040594    -0.091720   \n",
       "3   -0.043500    0.150232   -0.094052    -0.091896    -0.067998    -0.044945   \n",
       "4   -0.002303    0.097203   -0.041101    -0.062483    -0.059103    -0.107579   \n",
       "5    0.023726   -0.084342    0.012554    -0.168494    -0.097386    -0.081487   \n",
       "6    0.059421   -0.017267   -0.067476    -0.153824    -0.142285     0.099692   \n",
       "7   -0.052565    0.046599   -0.067784    -0.112384    -0.070892    -0.022191   \n",
       "8    0.007995    0.068928   -0.043145    -0.151496    -0.129461    -0.063685   \n",
       "9   -0.036615    0.175278   -0.085980    -0.147543    -0.099325     0.055207   \n",
       "\n",
       "   Component13  Component14  Component15  Component16  Component17  \\\n",
       "0    -0.055906    -0.107747     0.163747    -0.073270     0.097261   \n",
       "1     0.108397    -0.119072     0.298776    -0.120180    -0.013808   \n",
       "2     0.016219     0.024631     0.102074    -0.005710    -0.029763   \n",
       "3    -0.012982    -0.010849     0.072978    -0.041215     0.052298   \n",
       "4    -0.044388     0.004592    -0.010163    -0.051874     0.005546   \n",
       "5     0.092587    -0.126993     0.128796     0.014674     0.144005   \n",
       "6    -0.024039    -0.157745     0.087492    -0.010780     0.217738   \n",
       "7    -0.009099     0.002649     0.072151     0.024153     0.100830   \n",
       "8     0.084314    -0.146427     0.097307    -0.110862     0.070262   \n",
       "9    -0.115674    -0.055328     0.014914    -0.039102     0.144376   \n",
       "\n",
       "   Component18  Component19  Component20  Component21  Component22  \\\n",
       "0     0.001194     0.183316    -0.056198    -0.026070    -0.010077   \n",
       "1    -0.053277     0.283966    -0.279235     0.065307    -0.146740   \n",
       "2    -0.192161     0.161236    -0.206023     0.042199    -0.006985   \n",
       "3    -0.030508     0.154652    -0.102043    -0.058035     0.039161   \n",
       "4    -0.097045     0.125011    -0.062713    -0.113314     0.129667   \n",
       "5    -0.100702     0.104164    -0.003518     0.016678    -0.114431   \n",
       "6     0.010204     0.195356    -0.110219    -0.039211    -0.151496   \n",
       "7    -0.016991     0.137095    -0.060045    -0.055566    -0.062975   \n",
       "8    -0.228543     0.253672    -0.226056     0.094342    -0.148660   \n",
       "9    -0.045969     0.288825    -0.270984    -0.107616    -0.108731   \n",
       "\n",
       "   Component23  Component24  Component25  Component26  Component27  \\\n",
       "0     0.057767    -0.010817     0.034083     0.079074    -0.017282   \n",
       "1     0.032129    -0.095018    -0.030789     0.191273    -0.109182   \n",
       "2     0.001805     0.109591     0.070431     0.127238     0.021463   \n",
       "3     0.061158    -0.029621     0.043033     0.075030     0.043423   \n",
       "4     0.014687     0.111240     0.009607     0.124755     0.129615   \n",
       "5    -0.102848     0.137504     0.035710     0.023495     0.054907   \n",
       "6     0.042243    -0.061639     0.037335     0.105429    -0.062406   \n",
       "7     0.065279     0.012828     0.007876     0.055051    -0.023394   \n",
       "8     0.043638    -0.059827     0.138846     0.031970    -0.100223   \n",
       "9     0.175881    -0.109261     0.068311     0.061233    -0.067302   \n",
       "\n",
       "   Component28  Component29  Component30  Component31  Component32  \n",
       "0     0.132538    -0.119630     0.032354    -0.074276    -0.039071  \n",
       "1     0.206062    -0.282896     0.099974     0.008139    -0.049811  \n",
       "2     0.101687    -0.088881     0.006124    -0.071933    -0.029166  \n",
       "3     0.091814    -0.109135     0.073924    -0.042635    -0.049726  \n",
       "4     0.046244    -0.049326     0.012907    -0.077644    -0.079021  \n",
       "5     0.000656    -0.063511    -0.092066    -0.103760    -0.056617  \n",
       "6     0.053244    -0.139738     0.043533    -0.021184     0.059896  \n",
       "7     0.081747    -0.114850     0.027080    -0.057619    -0.033613  \n",
       "8     0.077079    -0.133173     0.042750     0.007902    -0.004513  \n",
       "9     0.151313    -0.097409     0.130504     0.031902     0.043764  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "viewinfo(to_use)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "25e8416b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T10:55:52.622517Z",
     "iopub.status.busy": "2022-10-06T10:55:52.622163Z",
     "iopub.status.idle": "2022-10-06T10:56:27.920081Z",
     "shell.execute_reply": "2022-10-06T10:56:27.919002Z"
    },
    "id": "25e8416b",
    "papermill": {
     "duration": 35.413447,
     "end_time": "2022-10-06T10:56:27.922501",
     "exception": false,
     "start_time": "2022-10-06T10:55:52.509054",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "kmeans6 = KMeans(n_clusters=6, random_state=RANDOM_STATE).fit(to_use.values)\n",
    "to_keep['labels'] = kmeans6.labels_"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f84920f9",
   "metadata": {},
   "source": [
    "## Deployment CSV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bbb99124",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T10:56:28.137420Z",
     "iopub.status.busy": "2022-10-06T10:56:28.136886Z",
     "iopub.status.idle": "2022-10-06T11:04:41.134445Z",
     "shell.execute_reply": "2022-10-06T11:04:41.132844Z"
    },
    "id": "bbb99124",
    "papermill": {
     "duration": 493.296727,
     "end_time": "2022-10-06T11:04:41.326077",
     "exception": false,
     "start_time": "2022-10-06T10:56:28.029350",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "to_keep.to_csv('../Data/path_group.csv', index=False)\n",
    "print(\"Converted!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9451a89b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T11:04:41.543807Z",
     "iopub.status.busy": "2022-10-06T11:04:41.543144Z",
     "iopub.status.idle": "2022-10-06T11:20:03.846249Z",
     "shell.execute_reply": "2022-10-06T11:20:03.845286Z"
    },
    "id": "9451a89b",
    "outputId": "93fff273-c8ff-4da6-95b9-203a10d63dca",
    "papermill": {
     "duration": 922.524735,
     "end_time": "2022-10-06T11:20:03.958173",
     "exception": false,
     "start_time": "2022-10-06T11:04:41.433438",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cluster numbers 2 -> inertia: 72442.64062\n",
      "cluster numbers 3 -> inertia: 65321.32031\n",
      "cluster numbers 4 -> inertia: 60404.44531\n",
      "cluster numbers 5 -> inertia: 56752.07031\n",
      "cluster numbers 6 -> inertia: 53963.31250\n",
      "cluster numbers 7 -> inertia: 51792.55469\n",
      "cluster numbers 8 -> inertia: 49936.51953\n",
      "cluster numbers 9 -> inertia: 48531.48438\n",
      "cluster numbers 10 -> inertia: 47267.90234\n",
      "cluster numbers 11 -> inertia: 46156.02734\n",
      "cluster numbers 12 -> inertia: 45181.12891\n",
      "cluster numbers 13 -> inertia: 44298.90625\n",
      "cluster numbers 14 -> inertia: 43535.37109\n",
      "cluster numbers 15 -> inertia: 42814.60938\n",
      "cluster numbers 16 -> inertia: 42141.15625\n",
      "cluster numbers 17 -> inertia: 41496.84375\n",
      "cluster numbers 18 -> inertia: 40880.20312\n",
      "cluster numbers 19 -> inertia: 40294.27734\n",
      "cluster numbers 20 -> inertia: 39792.85156\n",
      "cluster numbers 21 -> inertia: 39333.74609\n",
      "cluster numbers 22 -> inertia: 38900.62891\n",
      "cluster numbers 23 -> inertia: 38473.44922\n",
      "cluster numbers 24 -> inertia: 38079.77734\n",
      "cluster numbers 25 -> inertia: 37714.76562\n",
      "cluster numbers 26 -> inertia: 37374.13281\n",
      "cluster numbers 27 -> inertia: 37043.49609\n",
      "cluster numbers 28 -> inertia: 36701.76953\n",
      "cluster numbers 29 -> inertia: 36360.78906\n",
      "cluster numbers 30 -> inertia: 36050.64062\n",
      "cluster numbers 31 -> inertia: 35810.71875\n",
      "cluster numbers 32 -> inertia: 35474.89844\n"
     ]
    }
   ],
   "source": [
    "labels = []\n",
    "inertias = []\n",
    "N_CLUSTERS = vector_size\n",
    "for n_clusters in range(2, N_CLUSTERS+1):\n",
    "    kmeans = KMeans(n_clusters=n_clusters, random_state=RANDOM_STATE).fit(to_use.values)\n",
    "#     silhouette = silhouette_score(to_use, kmeans.labels_, metric=\"euclidean\", random_state=RANDOM_STATE)\n",
    "    inertia = kmeans.inertia_\n",
    "    inertias.append(inertia)\n",
    "    labels.append(kmeans.labels_)\n",
    "    print(f\"cluster numbers {n_clusters} -> inertia: {inertia:.5f}\")\n",
    "#     print(f\"cluster numbers {n_clusters} -> inertia: {inertia:.5f}\\tsilhouette: {silhouette:.5f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ccb6ecd5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:20:04.183836Z",
     "iopub.status.busy": "2022-10-06T11:20:04.182698Z",
     "iopub.status.idle": "2022-10-06T11:20:04.188404Z",
     "shell.execute_reply": "2022-10-06T11:20:04.187767Z"
    },
    "id": "ccb6ecd5",
    "papermill": {
     "duration": 0.12014,
     "end_time": "2022-10-06T11:20:04.190322",
     "exception": false,
     "start_time": "2022-10-06T11:20:04.070182",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "diffs = [np.inf]\n",
    "for i in range(1, len(inertias)):\n",
    "    diff = abs(inertias[i] - inertias[i-1])\n",
    "    diffs.append(diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7327a543",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:20:04.410822Z",
     "iopub.status.busy": "2022-10-06T11:20:04.410458Z",
     "iopub.status.idle": "2022-10-06T11:20:04.422406Z",
     "shell.execute_reply": "2022-10-06T11:20:04.421769Z"
    },
    "id": "7327a543",
    "papermill": {
     "duration": 0.123311,
     "end_time": "2022-10-06T11:20:04.424114",
     "exception": false,
     "start_time": "2022-10-06T11:20:04.300803",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "scores = pd.DataFrame(columns=['n_clusters', 'inertias', 'diffs'])\n",
    "scores['n_clusters'] = np.arange(2, N_CLUSTERS+1)\n",
    "scores['inertias'] = inertias\n",
    "scores['diffs'] = diffs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a2325cd1",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2022-10-06T11:20:04.643742Z",
     "iopub.status.busy": "2022-10-06T11:20:04.643130Z",
     "iopub.status.idle": "2022-10-06T11:20:05.388057Z",
     "shell.execute_reply": "2022-10-06T11:20:05.387104Z"
    },
    "id": "a2325cd1",
    "outputId": "aaaf759b-5a51-46b1-8561-38f22f557b87",
    "papermill": {
     "duration": 0.856606,
     "end_time": "2022-10-06T11:20:05.390259",
     "exception": false,
     "start_time": "2022-10-06T11:20:04.533653",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 6))\n",
    "\n",
    "plt.subplot(121)\n",
    "plt.plot(inertias)\n",
    "plt.grid(True)\n",
    "plt.title(\"Inertia\")\n",
    "\n",
    "plt.subplot(122)\n",
    "plt.plot(diffs)\n",
    "# plt.plot([0, 7], [diffs[5], diffs[5]], ':k')\n",
    "# plt.plot([7, 7], [0, diffs[5]], ':k')\n",
    "# plt.plot([0, 8], [diffs[6], diffs[6]], ':k')\n",
    "# plt.plot([8, 8], [0, diffs[6]], ':k')\n",
    "plt.grid(True)\n",
    "# plt.xlim([2, vector_size+1])\n",
    "plt.title(\"Difference in inertia\")\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f1982be",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7f1982be",
    "outputId": "5e224938-283d-4694-94a0-13f920ae3923",
    "papermill": {
     "duration": 0.109649,
     "end_time": "2022-10-06T11:20:05.665779",
     "exception": false,
     "start_time": "2022-10-06T11:20:05.556130",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "n_clusters        7.000000\n",
       "inertias      51792.554688\n",
       "diffs          2170.757812\n",
       "Name: 5, dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores.iloc[5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bf67c0cc",
   "metadata": {
    "id": "bf67c0cc",
    "papermill": {
     "duration": 0.109475,
     "end_time": "2022-10-06T11:20:05.885486",
     "exception": false,
     "start_time": "2022-10-06T11:20:05.776011",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "provenance": []
  },
  "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": "3.9.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 2209.682242,
   "end_time": "2022-10-06T11:20:08.125049",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2022-10-06T10:43:18.442807",
   "version": "2.3.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}