InstacartFeatureXG.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn import metrics, model_selection\n",
    "color = sns.color_palette()\n",
    "%matplotlib inline\n",
    "pd.options.mode.chained_assignment = None  # default='warn'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Changing current working directory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/rishikadawkar/Desktop/ADS /Instacart data\n",
      "/Users/rishikadawkar/Desktop/ADS /Instacart data\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "cwd = os.getcwd()\n",
    "print (cwd)\n",
    "%cd \"/Users/rishikadawkar/Desktop/ADS /Instacart data\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let's get the list of all the input files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AprioriInstacart.ipynb\n",
      "EDA.ipynb\n",
      "InstacartFeatureXG.ipynb\n",
      "NLTK.ipynb\n",
      "SamXG.ipynb\n",
      "SamXG2.ipynb\n",
      "aisles.csv\n",
      "clusterizat.ipynb\n",
      "departments.csv\n",
      "mllibSpark.ipynb\n",
      "order_products__prior.csv\n",
      "order_products__train.csv\n",
      "orders.csv\n",
      "products.csv\n",
      "sample_submission.csv\n",
      "script.py\n",
      "script1234.ipynb\n",
      "script7890.ipynb\n",
      "submission.csv\n",
      "trial_with_other_algo.ipynb\n",
      "xgb_starter_3450.csv\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from subprocess import check_output\n",
    "print(check_output([\"ls\", \"../Instacart data\"]).decode(\"utf8\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Read Inputs into data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "order_products_train = pd.read_csv(\"../Instacart data/order_products__train.csv\")\n",
    "order_products_prior = pd.read_csv(\"../Instacart data/order_products__prior.csv\")\n",
    "orders = pd.read_csv(\"../Instacart data/orders.csv\")\n",
    "products = pd.read_csv(\"../Instacart data/products.csv\")\n",
    "aisles = pd.read_csv(\"../Instacart data/aisles.csv\")\n",
    "departments = pd.read_csv(\"../Instacart data/departments.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Firstly, let's explore order_products_train and order_products_prior files. These files specify which products were purchased in each order. More specifically, order_products__prior contains previous order contents for all customers and order_products_train contains the last orders for some customers only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The order_products_train size is :  (1384617, 4)\n",
      "The order_products_prior size is :  (32434489, 4)\n"
     ]
    }
   ],
   "source": [
    "print(\"The order_products_train size is : \", order_products_train.shape)\n",
    "print(\"The order_products_prior size is : \", order_products_prior.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's concatenate them in the same order_products dataframe. This will concatenate all orders from previous customers plus new customers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The order_products_all size is :  (33819106, 4)\n"
     ]
    }
   ],
   "source": [
    "order_products_all = pd.concat([order_products_train, order_products_prior], axis=0)\n",
    "\n",
    "print(\"The order_products_all size is : \", order_products_all.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Missing Data in all orders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Total Missing</th>\n",
       "      <th>Percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>reordered</th>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>add_to_cart_order</th>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>product_id</th>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>order_id</th>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Total Missing  Percent\n",
       "reordered                      0      0.0\n",
       "add_to_cart_order              0      0.0\n",
       "product_id                     0      0.0\n",
       "order_id                       0      0.0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total = order_products_all.isnull().sum().sort_values(ascending=False)\n",
    "percent = (order_products_all.isnull().sum()/order_products_all.isnull().count()).sort_values(ascending=False)\n",
    "missing_data = pd.concat([total, percent], axis=1, keys=['Total Missing', 'Percent'])\n",
    "missing_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no missing data in order_products_all dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reorder Frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikadawkar/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:1: FutureWarning: using a dict on a Series for aggregation\n",
      "is deprecated and will be removed in a future version\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reordered</th>\n",
       "      <th>Total_products</th>\n",
       "      <th>Ratios</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>13863746</td>\n",
       "      <td>0.409938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>19955360</td>\n",
       "      <td>0.590062</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   reordered  Total_products    Ratios\n",
       "0          0        13863746  0.409938\n",
       "1          1        19955360  0.590062"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grouped = order_products_all.groupby(\"reordered\")[\"product_id\"].aggregate({'Total_products': 'count'}).reset_index()\n",
    "grouped['Ratios'] = grouped[\"Total_products\"].apply(lambda x: x /grouped['Total_products'].sum())\n",
    "grouped"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "59 % of ordered products are previously ordered by customers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWUAAAHdCAYAAADFHGijAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1119acac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "grouped  = grouped.groupby(['reordered']).sum()['Total_products'].sort_values(ascending=False)\n",
    "\n",
    "sns.set_style('whitegrid')\n",
    "f, ax = plt.subplots(figsize=(5, 8))\n",
    "sns.barplot(grouped.index, grouped.values, palette='RdBu_r')\n",
    "plt.ylabel('Number of Products', fontsize=13)\n",
    "plt.xlabel('Reordered or Not Reordered', fontsize=13)\n",
    "plt.ticklabel_format(style='plain', axis='y')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# To predict which previously purchased products will be in a user’s next order"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 1 We will start with reading the orders file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data_path = \"../Instacart data/\"\n",
    "orders_df = pd.read_csv(data_path + \"orders.csv\", usecols=[\"order_id\",\"user_id\",\"order_number\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the objective is to predict which previously purchased products will be in next order, let us first get the list of all products purchased by the customer."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 2 Create data frame with old purchase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>add_to_cart_order</th>\n",
       "      <th>reordered</th>\n",
       "      <th>user_id</th>\n",
       "      <th>order_number</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>33120</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>202279</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>28985</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>202279</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>9327</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>202279</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>45918</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>202279</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>30035</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>202279</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  product_id  add_to_cart_order  reordered  user_id  order_number\n",
       "0         2       33120                  1          1   202279             3\n",
       "1         2       28985                  2          1   202279             3\n",
       "2         2        9327                  3          0   202279             3\n",
       "3         2       45918                  4          1   202279             3\n",
       "4         2       30035                  5          0   202279             3"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# read the prior order file #\n",
    "prior_df = pd.read_csv(data_path + \"order_products__prior.csv\")\n",
    "\n",
    "# merge with the orders file to get the user_id #\n",
    "prior_df = pd.merge(prior_df, orders_df, how=\"inner\", on=\"order_id\")\n",
    "prior_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 3 Create data frame with latest purchase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_latest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>59897</td>\n",
       "      <td>9755</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>59897</td>\n",
       "      <td>31487</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>59897</td>\n",
       "      <td>37510</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>59897</td>\n",
       "      <td>14576</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59897</td>\n",
       "      <td>22105</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  product_id  reordered_latest\n",
       "0    59897        9755                 1\n",
       "1    59897       31487                 0\n",
       "2    59897       37510                 1\n",
       "3    59897       14576                 1\n",
       "4    59897       22105                 0"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the products and reorder status of the latest purchase of each user #\n",
    "prior_grouped_df = prior_df.groupby(\"user_id\")[\"order_number\"].aggregate(\"max\").reset_index()\n",
    "prior_df_latest = pd.merge(prior_df, prior_grouped_df, how=\"inner\", on=[\"user_id\", \"order_number\"])\n",
    "prior_df_latest = prior_df_latest[[\"user_id\", \"product_id\", \"reordered\"]]\n",
    "prior_df_latest.columns = [\"user_id\", \"product_id\", \"reordered_latest\"]\n",
    "prior_df_latest.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 4. Get count of Number of reorders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>196</td>\n",
       "      <td>10</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>10258</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>10326</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>12427</td>\n",
       "      <td>10</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>13032</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  product_id  reordered_count  reordered_sum\n",
       "0        1         196               10              9\n",
       "1        1       10258                9              8\n",
       "2        1       10326                1              0\n",
       "3        1       12427               10              9\n",
       "4        1       13032                3              2"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the count of each product and number of reorders by the customer #\n",
    "prior_df_agg = prior_df.groupby([\"user_id\",\"product_id\"])[\"reordered\"].aggregate([\"count\", \"sum\"]).reset_index()\n",
    "prior_df_agg.columns = [\"user_id\", \"product_id\", \"reordered_count\", \"reordered_sum\"]\n",
    "prior_df_agg.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 5 merge the prior df with latest df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>196</td>\n",
       "      <td>10</td>\n",
       "      <td>9</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>10258</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>10326</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>12427</td>\n",
       "      <td>10</td>\n",
       "      <td>9</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>13032</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  product_id  reordered_count  reordered_sum  reordered_latest\n",
       "0        1         196               10              9               1.0\n",
       "1        1       10258                9              8               1.0\n",
       "2        1       10326                1              0               NaN\n",
       "3        1       12427               10              9               1.0\n",
       "4        1       13032                3              2               1.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prior_df_merge = pd.merge(prior_df_agg, prior_df_latest, how=\"left\", on=[\"user_id\",\"product_id\"])\n",
    "prior_df_merge.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read the train and test dataset and then merge with orders data to get the user_id for the corresponding order_id."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id\n",
       "0         1\n",
       "1         1\n",
       "2         1\n",
       "3         1\n",
       "4         1"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "orders_df.drop([\"order_number\"],axis=1,inplace=True)\n",
    "\n",
    "train_df = pd.read_csv(data_path + \"order_products__train.csv\", usecols=[\"order_id\"])\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 6 Get distinct order id in train_df. This will help in joining the data frame with orders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id\n",
       "0         1\n",
       "1        36\n",
       "2        38\n",
       "3        96\n",
       "4        98"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = train_df.groupby(\"order_id\").aggregate(\"count\").reset_index()\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "join with Orders data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36</td>\n",
       "      <td>79431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38</td>\n",
       "      <td>42756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>96</td>\n",
       "      <td>17227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>98</td>\n",
       "      <td>56463</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id\n",
       "0         1   112108\n",
       "1        36    79431\n",
       "2        38    42756\n",
       "3        96    17227\n",
       "4        98    56463"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = pd.merge(train_df, orders_df, how=\"inner\", on=\"order_id\")\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>34</td>\n",
       "      <td>35220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>137</td>\n",
       "      <td>187107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>182</td>\n",
       "      <td>115892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>257</td>\n",
       "      <td>35581</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id\n",
       "0        17    36855\n",
       "1        34    35220\n",
       "2       137   187107\n",
       "3       182   115892\n",
       "4       257    35581"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.read_csv(data_path + \"sample_submission.csv\", usecols=[\"order_id\"])\n",
    "\n",
    "test_df = pd.merge(test_df, orders_df, how=\"inner\", on=\"order_id\")\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 7 : Prepare test data frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(131209, 2) (75000, 2)\n"
     ]
    }
   ],
   "source": [
    "print(train_df.shape, test_df.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 8: Merge the train and test data with prior_df to get the products purchased previously by the customer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>5707</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>11109</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>14947</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>22035</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "0         1   112108        2067                1              0   \n",
       "1         1   112108        5707                2              1   \n",
       "2         1   112108       11109                2              1   \n",
       "3         1   112108       14947                3              2   \n",
       "4         1   112108       22035                2              1   \n",
       "\n",
       "   reordered_latest  \n",
       "0               NaN  \n",
       "1               1.0  \n",
       "2               NaN  \n",
       "3               1.0  \n",
       "4               NaN  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = pd.merge(train_df, prior_df_merge, how=\"inner\", on=\"user_id\")\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>1283</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>6291</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>7035</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>11494</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>13107</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "0        17    36855        1283                1              0   \n",
       "1        17    36855        6291                1              0   \n",
       "2        17    36855        7035                1              0   \n",
       "3        17    36855       11494                1              0   \n",
       "4        17    36855       13107                3              2   \n",
       "\n",
       "   reordered_latest  \n",
       "0               NaN  \n",
       "1               NaN  \n",
       "2               NaN  \n",
       "3               NaN  \n",
       "4               1.0  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.merge(test_df, prior_df_merge, how=\"inner\", on=\"user_id\")\n",
    "#del prior_df, prior_grouped_df, prior_df_latest\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>product_id</th>\n",
       "      <th>aisle_id</th>\n",
       "      <th>department_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>61</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>104</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>94</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>38</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   product_id  aisle_id  department_id\n",
       "0           1        61             19\n",
       "1           2       104             13\n",
       "2           3        94              7\n",
       "3           4        38              1\n",
       "4           5         5             13"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "products_df = pd.read_csv(data_path + \"products.csv\", usecols=[\"product_id\", \"aisle_id\", \"department_id\"])\n",
    "products_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 9 : Add more features to training data from Products data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "      <th>aisle_id</th>\n",
       "      <th>department_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>96</td>\n",
       "      <td>17227</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3243</td>\n",
       "      <td>206024</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12950</td>\n",
       "      <td>11456</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17683</td>\n",
       "      <td>177724</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "0         1   112108        2067                1              0   \n",
       "1        96    17227        2067                1              0   \n",
       "2      3243   206024        2067                1              0   \n",
       "3     12950    11456        2067                1              0   \n",
       "4     17683   177724        2067                1              0   \n",
       "\n",
       "   reordered_latest  aisle_id  department_id  \n",
       "0               NaN         3             19  \n",
       "1               NaN         3             19  \n",
       "2               NaN         3             19  \n",
       "3               0.0         3             19  \n",
       "4               NaN         3             19  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = pd.merge(train_df, products_df, how=\"inner\", on=\"product_id\")\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "      <th>aisle_id</th>\n",
       "      <th>department_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17</td>\n",
       "      <td>36855</td>\n",
       "      <td>1283</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>657743</td>\n",
       "      <td>16994</td>\n",
       "      <td>1283</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>834223</td>\n",
       "      <td>92628</td>\n",
       "      <td>1283</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1446886</td>\n",
       "      <td>75870</td>\n",
       "      <td>1283</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1518981</td>\n",
       "      <td>118458</td>\n",
       "      <td>1283</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "0        17    36855        1283                1              0   \n",
       "1    657743    16994        1283                1              0   \n",
       "2    834223    92628        1283                1              0   \n",
       "3   1446886    75870        1283                1              0   \n",
       "4   1518981   118458        1283                3              2   \n",
       "\n",
       "   reordered_latest  aisle_id  department_id  \n",
       "0               NaN       100             21  \n",
       "1               NaN       100             21  \n",
       "2               NaN       100             21  \n",
       "3               NaN       100             21  \n",
       "4               NaN       100             21  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.merge(test_df, products_df, how=\"inner\", on=\"product_id\")\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have all the products that has been purchased previously by the customer along with some characteristics / featrures. So we can use the train dataset to populate the target variable i.e., whether the product has been reordered in the next order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(8474661, 9)\n"
     ]
    }
   ],
   "source": [
    "train_y_df = pd.read_csv(data_path + \"order_products__train.csv\", usecols=[\"order_id\", \"product_id\", \"reordered\"])\n",
    "train_y_df = pd.merge(train_y_df, orders_df, how=\"inner\", on=\"order_id\")\n",
    "train_y_df = train_y_df[[\"user_id\", \"product_id\", \"reordered\"]]\n",
    "train_df = pd.merge(train_df, train_y_df, how=\"left\", on=[\"user_id\", \"product_id\"])\n",
    "train_df[\"reordered\"].fillna(0, inplace=True)\n",
    "print(train_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from numpy import loadtxt\n",
    "import xgboost as xgb\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X=train_df.drop('reordered',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "      <th>aisle_id</th>\n",
       "      <th>department_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>112108</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>96</td>\n",
       "      <td>17227</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3243</td>\n",
       "      <td>206024</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12950</td>\n",
       "      <td>11456</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17683</td>\n",
       "      <td>177724</td>\n",
       "      <td>2067</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "0         1   112108        2067                1              0   \n",
       "1        96    17227        2067                1              0   \n",
       "2      3243   206024        2067                1              0   \n",
       "3     12950    11456        2067                1              0   \n",
       "4     17683   177724        2067                1              0   \n",
       "\n",
       "   reordered_latest  aisle_id  department_id  \n",
       "0               NaN         3             19  \n",
       "1               NaN         3             19  \n",
       "2               NaN         3             19  \n",
       "3               0.0         3             19  \n",
       "4               NaN         3             19  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y =train_df['reordered']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.0\n",
       "1    0.0\n",
       "2    0.0\n",
       "3    0.0\n",
       "4    0.0\n",
       "Name: reordered, dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 10: split data into train and test sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, random_state=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = XGBClassifier()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 11: Train the model using XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=1,\n",
       "       gamma=0, learning_rate=0.1, max_delta_step=0, max_depth=3,\n",
       "       min_child_weight=1, missing=None, n_estimators=100, nthread=-1,\n",
       "       objective='binary:logistic', reg_alpha=0, reg_lambda=1,\n",
       "       scale_pos_weight=1, seed=0, silent=True, subsample=1)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "predictions = [round(value) for value in y_pred]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "accuracy = accuracy_score(y_test, predictions)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Accuracy for the  XGBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 90.43%\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: %.2f%%\" % (accuracy * 100.0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(8474661, 6) (4833292, 6)\n"
     ]
    }
   ],
   "source": [
    "# target variable for train set #\n",
    "train_y = train_df.reordered.values\n",
    "\n",
    "# dataframe for test set predictions #\n",
    "out_df = test_df[[\"order_id\", \"product_id\"]]\n",
    "\n",
    "# drop the unnecessary columns #\n",
    "train_df = np.array(train_df.drop([\"order_id\", \"user_id\", \"reordered\"], axis=1))\n",
    "test_df = np.array(test_df.drop([\"order_id\", \"user_id\"], axis=1))\n",
    "print(train_df.shape, test_df.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 12: Create XGb function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def runXGB(train_X, train_y, test_X, test_y=None, feature_names=None, seed_val=0):\n",
    "        params = {}\n",
    "        params[\"objective\"] = \"binary:logistic\"\n",
    "        params['eval_metric'] = 'logloss'\n",
    "        params[\"eta\"] = 0.05\n",
    "        params[\"subsample\"] = 0.7\n",
    "        params[\"min_child_weight\"] = 10\n",
    "        params[\"colsample_bytree\"] = 0.7\n",
    "        params[\"max_depth\"] = 8\n",
    "        params[\"silent\"] = 1\n",
    "        params[\"seed\"] = seed_val\n",
    "        num_rounds = 100\n",
    "        plst = list(params.items())\n",
    "        xgtrain = xgb.DMatrix(train_X, label=train_y)\n",
    "\n",
    "        if test_y is not None:\n",
    "                xgtest = xgb.DMatrix(test_X, label=test_y)\n",
    "                watchlist = [ (xgtrain,'train'), (xgtest, 'test') ]\n",
    "                model = xgb.train(plst, xgtrain, num_rounds, watchlist, early_stopping_rounds=50, verbose_eval=10)\n",
    "        else:\n",
    "                xgtest = xgb.DMatrix(test_X)\n",
    "                model = xgb.train(plst, xgtrain, num_rounds)\n",
    "\n",
    "        pred_test_y = model.predict(xgtest)\n",
    "        return pred_test_y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 13: Run the XGB function on training np array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikadawkar/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: DeprecationWarning: \n",
      ".ix is deprecated. Please use\n",
      ".loc for label based indexing or\n",
      ".iloc for positional indexing\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated\n",
      "  # Remove the CWD from sys.path while we load stuff.\n"
     ]
    }
   ],
   "source": [
    "# run the xgboost model #\n",
    "pred = runXGB(train_df, train_y, test_df)\n",
    "#del train_df, test_df\n",
    "\n",
    "# use a cut-off value to get the predictions #\n",
    "cutoff = 0.2\n",
    "pred[pred>=cutoff] = 1\n",
    "pred[pred<cutoff] = 0\n",
    "out_df[\"Pred\"] = pred\n",
    "out_df = out_df.ix[out_df[\"Pred\"].astype('int')==1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# when there are more than 1 product, merge them to a single string #\n",
    "def merge_products(x):\n",
    "    return \" \".join(list(x.astype('str')))\n",
    "out_df = out_df.groupby(\"order_id\")[\"product_id\"].aggregate(merge_products).reset_index()\n",
    "out_df.columns = [\"order_id\", \"products\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Step 14: Load result in the output file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# read the sample csv file and populate the products from predictions #\n",
    "sub_df = pd.read_csv(data_path + \"sample_submission.csv\", usecols=[\"order_id\"])\n",
    "sub_df = pd.merge(sub_df, out_df, how=\"left\", on=\"order_id\")\n",
    "\n",
    "# when there are no predictions use \"None\" #\n",
    "sub_df[\"products\"].fillna(\"None\", inplace=True)\n",
    "sub_df.to_csv(\"xgb_starter_3450.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>reordered_count</th>\n",
       "      <th>reordered_sum</th>\n",
       "      <th>reordered_latest</th>\n",
       "      <th>aisle_id</th>\n",
       "      <th>department_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3052321</th>\n",
       "      <td>2634088</td>\n",
       "      <td>11335</td>\n",
       "      <td>6873</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5273294</th>\n",
       "      <td>3379345</td>\n",
       "      <td>104957</td>\n",
       "      <td>15923</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>67</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6564246</th>\n",
       "      <td>1344587</td>\n",
       "      <td>174152</td>\n",
       "      <td>45692</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>107</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7705447</th>\n",
       "      <td>2206969</td>\n",
       "      <td>196469</td>\n",
       "      <td>31205</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6947171</th>\n",
       "      <td>1103780</td>\n",
       "      <td>57331</td>\n",
       "      <td>28599</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>130</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3706974</th>\n",
       "      <td>2174683</td>\n",
       "      <td>125292</td>\n",
       "      <td>12845</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>117</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8344595</th>\n",
       "      <td>1026666</td>\n",
       "      <td>56033</td>\n",
       "      <td>45362</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>51</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5605328</th>\n",
       "      <td>2860146</td>\n",
       "      <td>80015</td>\n",
       "      <td>43460</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>123</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5248902</th>\n",
       "      <td>15763</td>\n",
       "      <td>71816</td>\n",
       "      <td>19863</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>107</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>462416</th>\n",
       "      <td>232693</td>\n",
       "      <td>77967</td>\n",
       "      <td>41220</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3534830</th>\n",
       "      <td>1949599</td>\n",
       "      <td>126310</td>\n",
       "      <td>22455</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>61</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152552</th>\n",
       "      <td>2345085</td>\n",
       "      <td>14486</td>\n",
       "      <td>16759</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7075812</th>\n",
       "      <td>546353</td>\n",
       "      <td>10292</td>\n",
       "      <td>5640</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8350743</th>\n",
       "      <td>1529794</td>\n",
       "      <td>45915</td>\n",
       "      <td>33669</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>47</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>932819</th>\n",
       "      <td>1326500</td>\n",
       "      <td>151546</td>\n",
       "      <td>8021</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>54</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>634424</th>\n",
       "      <td>975625</td>\n",
       "      <td>96703</td>\n",
       "      <td>2124</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4886941</th>\n",
       "      <td>605046</td>\n",
       "      <td>128734</td>\n",
       "      <td>21783</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>853923</th>\n",
       "      <td>2935574</td>\n",
       "      <td>80460</td>\n",
       "      <td>42265</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>123</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>419655</th>\n",
       "      <td>2998179</td>\n",
       "      <td>194826</td>\n",
       "      <td>35939</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7392080</th>\n",
       "      <td>1346708</td>\n",
       "      <td>65077</td>\n",
       "      <td>37890</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120382</th>\n",
       "      <td>191662</td>\n",
       "      <td>176974</td>\n",
       "      <td>9339</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>106</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8376309</th>\n",
       "      <td>1501971</td>\n",
       "      <td>50250</td>\n",
       "      <td>10920</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3979156</th>\n",
       "      <td>1667567</td>\n",
       "      <td>49263</td>\n",
       "      <td>7632</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>72</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5171212</th>\n",
       "      <td>2816793</td>\n",
       "      <td>192902</td>\n",
       "      <td>5120</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>91</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7198817</th>\n",
       "      <td>724567</td>\n",
       "      <td>8537</td>\n",
       "      <td>39415</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>104</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3374561</th>\n",
       "      <td>2242001</td>\n",
       "      <td>91387</td>\n",
       "      <td>13212</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>104</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3166358</th>\n",
       "      <td>3237266</td>\n",
       "      <td>139704</td>\n",
       "      <td>32433</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>116</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006024</th>\n",
       "      <td>3318883</td>\n",
       "      <td>162705</td>\n",
       "      <td>42736</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2162102</th>\n",
       "      <td>2933314</td>\n",
       "      <td>165923</td>\n",
       "      <td>13198</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>122</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2056046</th>\n",
       "      <td>685749</td>\n",
       "      <td>150118</td>\n",
       "      <td>10957</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>77</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>163256</th>\n",
       "      <td>1975041</td>\n",
       "      <td>101364</td>\n",
       "      <td>19048</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4448386</th>\n",
       "      <td>1185135</td>\n",
       "      <td>11392</td>\n",
       "      <td>35018</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8314439</th>\n",
       "      <td>1323235</td>\n",
       "      <td>103227</td>\n",
       "      <td>18311</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>127</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6064686</th>\n",
       "      <td>1926655</td>\n",
       "      <td>17049</td>\n",
       "      <td>4289</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8276278</th>\n",
       "      <td>2783277</td>\n",
       "      <td>37625</td>\n",
       "      <td>32091</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1867829</th>\n",
       "      <td>446443</td>\n",
       "      <td>96660</td>\n",
       "      <td>14678</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>116</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1210708</th>\n",
       "      <td>1278001</td>\n",
       "      <td>39418</td>\n",
       "      <td>35042</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6339780</th>\n",
       "      <td>2317913</td>\n",
       "      <td>182465</td>\n",
       "      <td>8223</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1703958</th>\n",
       "      <td>486212</td>\n",
       "      <td>54577</td>\n",
       "      <td>37710</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>125</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2897807</th>\n",
       "      <td>753264</td>\n",
       "      <td>63056</td>\n",
       "      <td>22882</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>407478</th>\n",
       "      <td>2791029</td>\n",
       "      <td>143348</td>\n",
       "      <td>34126</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4176911</th>\n",
       "      <td>2444859</td>\n",
       "      <td>41833</td>\n",
       "      <td>48871</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>81</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3111695</th>\n",
       "      <td>1322715</td>\n",
       "      <td>186249</td>\n",
       "      <td>23165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3305494</th>\n",
       "      <td>1620270</td>\n",
       "      <td>23331</td>\n",
       "      <td>10644</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7593250</th>\n",
       "      <td>2861237</td>\n",
       "      <td>154062</td>\n",
       "      <td>35865</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>69</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3114083</th>\n",
       "      <td>2663723</td>\n",
       "      <td>86688</td>\n",
       "      <td>23165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1880042</th>\n",
       "      <td>3416466</td>\n",
       "      <td>54088</td>\n",
       "      <td>21267</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>112</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8184488</th>\n",
       "      <td>1349969</td>\n",
       "      <td>50097</td>\n",
       "      <td>961</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7550567</th>\n",
       "      <td>2335041</td>\n",
       "      <td>154775</td>\n",
       "      <td>10131</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>120</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3160939</th>\n",
       "      <td>3174825</td>\n",
       "      <td>122942</td>\n",
       "      <td>31066</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>85</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3747474</th>\n",
       "      <td>1598025</td>\n",
       "      <td>123149</td>\n",
       "      <td>6348</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5865723</th>\n",
       "      <td>2756562</td>\n",
       "      <td>174406</td>\n",
       "      <td>19204</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>121</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1442064</th>\n",
       "      <td>633045</td>\n",
       "      <td>46514</td>\n",
       "      <td>46294</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3962434</th>\n",
       "      <td>3067993</td>\n",
       "      <td>66478</td>\n",
       "      <td>47521</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>78</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8460265</th>\n",
       "      <td>968438</td>\n",
       "      <td>150502</td>\n",
       "      <td>23632</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>109</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1965286</th>\n",
       "      <td>1099888</td>\n",
       "      <td>79357</td>\n",
       "      <td>8277</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4473121</th>\n",
       "      <td>256984</td>\n",
       "      <td>62140</td>\n",
       "      <td>42959</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>74</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8158334</th>\n",
       "      <td>355337</td>\n",
       "      <td>70111</td>\n",
       "      <td>24341</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6278263</th>\n",
       "      <td>1455864</td>\n",
       "      <td>170500</td>\n",
       "      <td>11097</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>30</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4360443</th>\n",
       "      <td>692804</td>\n",
       "      <td>192171</td>\n",
       "      <td>47788</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         order_id  user_id  product_id  reordered_count  reordered_sum  \\\n",
       "3052321   2634088    11335        6873                1              0   \n",
       "5273294   3379345   104957       15923                1              0   \n",
       "6564246   1344587   174152       45692                1              0   \n",
       "7705447   2206969   196469       31205                1              0   \n",
       "6947171   1103780    57331       28599                1              0   \n",
       "3706974   2174683   125292       12845                1              0   \n",
       "8344595   1026666    56033       45362                2              1   \n",
       "5605328   2860146    80015       43460                8              7   \n",
       "5248902     15763    71816       19863                5              4   \n",
       "462416     232693    77967       41220                1              0   \n",
       "3534830   1949599   126310       22455                4              3   \n",
       "152552    2345085    14486       16759                2              1   \n",
       "7075812    546353    10292        5640                4              3   \n",
       "8350743   1529794    45915       33669                2              1   \n",
       "932819    1326500   151546        8021                6              5   \n",
       "634424     975625    96703        2124                5              4   \n",
       "4886941    605046   128734       21783                5              4   \n",
       "853923    2935574    80460       42265                1              0   \n",
       "419655    2998179   194826       35939                3              2   \n",
       "7392080   1346708    65077       37890                2              1   \n",
       "120382     191662   176974        9339                2              1   \n",
       "8376309   1501971    50250       10920                3              2   \n",
       "3979156   1667567    49263        7632                1              0   \n",
       "5171212   2816793   192902        5120                5              4   \n",
       "7198817    724567     8537       39415                2              1   \n",
       "3374561   2242001    91387       13212                1              0   \n",
       "3166358   3237266   139704       32433                1              0   \n",
       "2006024   3318883   162705       42736                1              0   \n",
       "2162102   2933314   165923       13198                1              0   \n",
       "2056046    685749   150118       10957                1              0   \n",
       "...           ...      ...         ...              ...            ...   \n",
       "163256    1975041   101364       19048                1              0   \n",
       "4448386   1185135    11392       35018                1              0   \n",
       "8314439   1323235   103227       18311                1              0   \n",
       "6064686   1926655    17049        4289                1              0   \n",
       "8276278   2783277    37625       32091                1              0   \n",
       "1867829    446443    96660       14678                1              0   \n",
       "1210708   1278001    39418       35042                1              0   \n",
       "6339780   2317913   182465        8223                1              0   \n",
       "1703958    486212    54577       37710                2              1   \n",
       "2897807    753264    63056       22882                1              0   \n",
       "407478    2791029   143348       34126                3              2   \n",
       "4176911   2444859    41833       48871                3              2   \n",
       "3111695   1322715   186249       23165                1              0   \n",
       "3305494   1620270    23331       10644                1              0   \n",
       "7593250   2861237   154062       35865                4              3   \n",
       "3114083   2663723    86688       23165                1              0   \n",
       "1880042   3416466    54088       21267                1              0   \n",
       "8184488   1349969    50097         961                1              0   \n",
       "7550567   2335041   154775       10131                1              0   \n",
       "3160939   3174825   122942       31066                1              0   \n",
       "3747474   1598025   123149        6348                3              2   \n",
       "5865723   2756562   174406       19204                2              1   \n",
       "1442064    633045    46514       46294                1              0   \n",
       "3962434   3067993    66478       47521                2              1   \n",
       "8460265    968438   150502       23632                1              0   \n",
       "1965286   1099888    79357        8277                9              8   \n",
       "4473121    256984    62140       42959                1              0   \n",
       "8158334    355337    70111       24341                1              0   \n",
       "6278263   1455864   170500       11097                1              0   \n",
       "4360443    692804   192171       47788                5              4   \n",
       "\n",
       "         reordered_latest  aisle_id  department_id  \n",
       "3052321               0.0        83              4  \n",
       "5273294               0.0        67             20  \n",
       "6564246               0.0       107             19  \n",
       "7705447               0.0        26              7  \n",
       "6947171               0.0       130             14  \n",
       "3706974               0.0       117             19  \n",
       "8344595               0.0        51             13  \n",
       "5605328               0.0       123              4  \n",
       "5248902               0.0       107             19  \n",
       "462416                0.0        83              4  \n",
       "3534830               0.0        61             19  \n",
       "152552                0.0        83              4  \n",
       "7075812               0.0        91             16  \n",
       "8350743               0.0        47             11  \n",
       "932819                0.0        54             17  \n",
       "634424                0.0        12              9  \n",
       "4886941               0.0        37              1  \n",
       "853923                0.0       123              4  \n",
       "419655                0.0        83              4  \n",
       "7392080               0.0         1             20  \n",
       "120382                0.0       106             12  \n",
       "8376309               0.0        21             16  \n",
       "3979156               0.0        72             13  \n",
       "5171212               1.0        91             16  \n",
       "7198817               0.0       104             13  \n",
       "3374561               0.0       104             13  \n",
       "3166358               0.0       116              1  \n",
       "2006024               0.0        36             16  \n",
       "2162102               0.0       122             12  \n",
       "2056046               0.0        77              7  \n",
       "...                   ...       ...            ...  \n",
       "163256                0.0        83              4  \n",
       "4448386               0.0        11             11  \n",
       "8314439               0.0       127             11  \n",
       "6064686               0.0        31              7  \n",
       "8276278               0.0         3             19  \n",
       "1867829               0.0       116              1  \n",
       "1210708               0.0        59             15  \n",
       "6339780               0.0        79              1  \n",
       "1703958               0.0       125             19  \n",
       "2897807               0.0        21             16  \n",
       "407478                0.0        16              4  \n",
       "4176911               1.0        81             15  \n",
       "3111695               0.0        83              4  \n",
       "3305494               0.0        19             13  \n",
       "7593250               0.0        69             15  \n",
       "3114083               0.0        83              4  \n",
       "1880042               0.0       112              3  \n",
       "8184488               0.0        13             20  \n",
       "7550567               0.0       120             16  \n",
       "3160939               0.0        85             17  \n",
       "3747474               0.0        21             16  \n",
       "5865723               0.0       121             14  \n",
       "1442064               0.0        16              4  \n",
       "3962434               0.0        78             19  \n",
       "8460265               0.0       109             11  \n",
       "1965286               0.0        24              4  \n",
       "4473121               0.0        74             17  \n",
       "8158334               0.0        31              7  \n",
       "6278263               0.0        30              6  \n",
       "4360443               0.0        24              4  \n",
       "\n",
       "[1000 rows x 8 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['reordered_latest'] = X_train['reordered_latest'].replace(np.nan, 0)\n",
    "X_test['reordered_latest'] = X_test['reordered_latest'].replace(np.nan, 0)\n",
    "X_test.head(1000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Logistic Regression model\n",
    "clf=(LogisticRegression(C=0.02))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=0.02, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fitting the model\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#predictions\n",
    "pred=clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.901983416522\n"
     ]
    }
   ],
   "source": [
    "#accuracy score of Logistic Regression Model\n",
    "print(accuracy_score(pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Randon Forest "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clfrf = RandomForestClassifier(max_features=\"log2\", max_depth=11, n_estimators=24,min_samples_split=1000, oob_score=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikadawkar/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:453: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/rishikadawkar/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:458: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=11, max_features='log2', max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=1, min_samples_split=1000,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators=24, n_jobs=1,\n",
       "            oob_score=True, random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clfrf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#predictions\n",
    "predrf=clfrf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.90456151115678496"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#accuracy score for the random forest model\n",
    "accuracy_score(predrf, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Light GBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import lightgbm as lgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_X = lgb.Dataset(X_train)\n",
    "train_y = lgb.Dataset(y_train)\n",
    "test_X = lgb.Dataset(X_test)\n",
    "test_y = lgb.Dataset(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model= lgb.LGBMClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "        learning_rate=0.1, max_depth=-1, min_child_samples=20,\n",
       "        min_child_weight=0.001, min_split_gain=0.0, n_estimators=100,\n",
       "        n_jobs=-1, num_leaves=31, objective=None, random_state=None,\n",
       "        reg_alpha=0.0, reg_lambda=0.0, silent=True, subsample=1.0,\n",
       "        subsample_for_bin=200000, subsample_freq=1)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred=model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.90463624372\n"
     ]
    }
   ],
   "source": [
    "#accuracy score of Light GBM\n",
    "print(accuracy_score(pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "The text in the document is from source Instacart Kaggle Competition is licensed under CC BY 3.0 https://creativecommons.org/licenses/by/3.0/us/\n",
    "\n",
    "The code in the document is by Onodera is licensed under the MIT License https://opensource.org/licenses/MIT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}