UNIC_Final.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# machine learning\n",
    "from sklearn.model_selection import train_test_split, KFold, cross_val_score, cross_val_predict\n",
    "from sklearn.svm import SVC, LinearSVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn import datasets, linear_model\n",
    "from sklearn.metrics import mean_squared_error, r2_score, classification_report, confusion_matrix\n",
    "from sklearn import preprocessing\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.metrics import roc_curve\n",
    "import shap\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "#Visualization \n",
    "import matplotlib.pyplot as plt \n",
    "plt.rc(\"font\", size=14)\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import datetime\n",
    "import seaborn as sns\n",
    "import math\n",
    "sns.set(style=\"white\")\n",
    "sns.set(style=\"whitegrid\", color_codes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading Bid Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Material Number</th>\n",
       "      <th>Product Family</th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Contract Status</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Date.1</th>\n",
       "      <th>Bid Submitted Month</th>\n",
       "      <th>...</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2018)</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2019)</th>\n",
       "      <th>Forecast</th>\n",
       "      <th>Launch Date</th>\n",
       "      <th>Launch Date Serial</th>\n",
       "      <th>Supply Quantity (Replenishment)</th>\n",
       "      <th>Market Share %</th>\n",
       "      <th>Number of Competitors</th>\n",
       "      <th>WAC Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51025.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>65617</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>NEW BUSINESS</td>\n",
       "      <td>2017-01-10 00:00:00</td>\n",
       "      <td>2017-01-01</td>\n",
       "      <td>42736.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>not available</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>905.0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51024.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>54765</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>NEW BUSINESS</td>\n",
       "      <td>2017-01-10 00:00:00</td>\n",
       "      <td>2017-01-01</td>\n",
       "      <td>42736.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>not available</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48912.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>121861</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>NEW BUSINESS</td>\n",
       "      <td>2017-01-10 00:00:00</td>\n",
       "      <td>2017-01-01</td>\n",
       "      <td>42736.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>not available</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1043.0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48914.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>101706</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>NEW BUSINESS</td>\n",
       "      <td>2017-01-10 00:00:00</td>\n",
       "      <td>2017-01-01</td>\n",
       "      <td>42736.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>not available</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2008.0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>41030.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1000813.0</td>\n",
       "      <td>59481.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>4190</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>NEW BUSINESS</td>\n",
       "      <td>2017-02-04 00:00:00</td>\n",
       "      <td>2017-01-01</td>\n",
       "      <td>42736.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>not available</td>\n",
       "      <td>2.569672</td>\n",
       "      <td>0.225163</td>\n",
       "      <td>522.0</td>\n",
       "      <td>2012-05-01</td>\n",
       "      <td>41030.0</td>\n",
       "      <td>1706.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer ID  Material Number  Product Family Annual units  Monthly units  \\\n",
       "0    1001517.0          51025.0           381.0        65617    5468.000000   \n",
       "1    1001517.0          51024.0           381.0        54765    4564.000000   \n",
       "2    1001517.0          48912.0           381.0       121861   10155.000000   \n",
       "3    1001517.0          48914.0           381.0       101706    8476.000000   \n",
       "4    1000813.0          59481.0           322.0         4190     349.166667   \n",
       "\n",
       "  Contract Status Customer Supply Start date Bid Submitted Date  \\\n",
       "0    NEW BUSINESS        2017-01-10 00:00:00         2017-01-01   \n",
       "1    NEW BUSINESS        2017-01-10 00:00:00         2017-01-01   \n",
       "2    NEW BUSINESS        2017-01-10 00:00:00         2017-01-01   \n",
       "3    NEW BUSINESS        2017-01-10 00:00:00         2017-01-01   \n",
       "4    NEW BUSINESS        2017-02-04 00:00:00         2017-01-01   \n",
       "\n",
       "   Bid Submitted Date.1  Bid Submitted Month  ...  \\\n",
       "0               42736.0                  1.0  ...   \n",
       "1               42736.0                  1.0  ...   \n",
       "2               42736.0                  1.0  ...   \n",
       "3               42736.0                  1.0  ...   \n",
       "4               42736.0                  1.0  ...   \n",
       "\n",
       "   Failure to Supply Rate (FTS) per unit  \\\n",
       "0                          not available   \n",
       "1                          not available   \n",
       "2                          not available   \n",
       "3                          not available   \n",
       "4                          not available   \n",
       "\n",
       "  Failure to Supply - Weighted Avg Exposure (2018)  \\\n",
       "0                                              NaN   \n",
       "1                                              NaN   \n",
       "2                                              NaN   \n",
       "3                                              NaN   \n",
       "4                                         2.569672   \n",
       "\n",
       "  Failure to Supply - Weighted Avg Exposure (2019) Forecast  Launch Date  \\\n",
       "0                                              NaN    905.0          NaT   \n",
       "1                                              NaN   1400.0          NaT   \n",
       "2                                              NaN   1043.0          NaT   \n",
       "3                                              NaN   2008.0          NaT   \n",
       "4                                         0.225163    522.0   2012-05-01   \n",
       "\n",
       "   Launch Date Serial  Supply Quantity (Replenishment) Market Share %  \\\n",
       "0                 0.0                              NaN            NaN   \n",
       "1                 0.0                              NaN            NaN   \n",
       "2                 0.0                              NaN            NaN   \n",
       "3             41030.0                              NaN            NaN   \n",
       "4             41030.0                           1706.0            NaN   \n",
       "\n",
       "   Number of Competitors  WAC Price  \n",
       "0                    NaN        NaN  \n",
       "1                    NaN        NaN  \n",
       "2                    NaN        NaN  \n",
       "3                    NaN        NaN  \n",
       "4                    NaN        NaN  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importing data set \n",
    "bid_data_df = pd.read_excel('Bid_Data_Expanded.xlsx',sheet_name=\"Bid Data\",header=0) \n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                                float64\n",
       "Material Number                                            float64\n",
       "Product Family                                             float64\n",
       "Annual units                                                object\n",
       "Monthly units                                              float64\n",
       "Contract Status                                             object\n",
       "Customer Supply Start date                                  object\n",
       "Bid Submitted Date                                  datetime64[ns]\n",
       "Bid Submitted Date.1                                       float64\n",
       "Bid Submitted Month                                        float64\n",
       "Bid Submitted Year                                         float64\n",
       "Apotex Response                                             object\n",
       "Customer Response                                           object\n",
       "Failure to Supply Rate (FTS) per unit                       object\n",
       "Failure to Supply - Weighted Avg Exposure (2018)           float64\n",
       "Failure to Supply - Weighted Avg Exposure (2019)           float64\n",
       "Forecast                                                   float64\n",
       "Launch Date                                         datetime64[ns]\n",
       "Launch Date Serial                                         float64\n",
       "Supply Quantity (Replenishment)                            float64\n",
       "Market Share %                                             float64\n",
       "Number of Competitors                                      float64\n",
       "WAC Price                                                  float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bid_data_df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Fixes\n",
    "There are several issues with the current data set that need to be addressed\n",
    "- Customer Supply Start Date needs to be fixed (solved)\n",
    "- Annual Units needs to be a float (solved)\n",
    "- Contract Status needs to be 1-hot encoded\n",
    "- Bid Submitted Date needs to be fixed (solved)\n",
    "- Bid Submitted Month needs to be 1-hot encoded (solved)\n",
    "- Apotex Response needs to be 1-hot encoded (solved)\n",
    "- Customer Response we need to drop the empty rows (solved)\n",
    "- Figure out what to do for FTS columns (solved)\n",
    "- Figure out what to do with other columns (solved)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove all the rows where the 'Customer Supply Start date' is NaN\n",
    "# All the other rows in this case are also NaN (probably some excel stuff)\n",
    "bid_data_df = bid_data_df.loc[~bid_data_df['Customer Supply Start date'].isnull()]\n",
    "# Convert all the 'Customer Supply Start date' values to timestamps\n",
    "bid_data_df['Customer Supply Start date'] = [\n",
    "    datetime.datetime.strptime(x.strip(), '%m/%d/%Y').timestamp() \n",
    "    if not isinstance(x, datetime.datetime) else x.timestamp() \n",
    "    for x in list(bid_data_df['Customer Supply Start date'].values)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop the 'Bid Submitted Date.1' column\n",
    "bid_data_df = bid_data_df.drop(columns=['Bid Submitted Date.1'])\n",
    "# Convert Bid Submitted Date to timestamp\n",
    "bid_data_df['Bid Submitted Date'] = list(\n",
    "    x.timestamp() for x in list(bid_data_df['Bid Submitted Date'])\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert Annual Units into floats\n",
    "bid_data_df['Annual units'] = [\n",
    "    int(x.strip().replace(',', '')) \n",
    "    if isinstance(x, str) else float(x)\n",
    "    for x in list(bid_data_df[\"Annual units\"].values)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'EXISTING BUSINESS', 'NEW BUSINESS'}\n"
     ]
    }
   ],
   "source": [
    "# Convert Contract Status entries to the same format (all upper)\n",
    "bid_data_df['Contract Status'] = [x.upper() for x in list(bid_data_df['Contract Status'].values)]\n",
    "print(set(bid_data_df['Contract Status'].values))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                                float64\n",
       "Material Number                                            float64\n",
       "Product Family                                             float64\n",
       "Annual units                                               float64\n",
       "Monthly units                                              float64\n",
       "Contract Status                                             object\n",
       "Customer Supply Start date                                 float64\n",
       "Bid Submitted Date                                         float64\n",
       "Bid Submitted Month                                        float64\n",
       "Bid Submitted Year                                         float64\n",
       "Apotex Response                                             object\n",
       "Customer Response                                           object\n",
       "Failure to Supply Rate (FTS) per unit                      float64\n",
       "Failure to Supply - Weighted Avg Exposure (2018)           float64\n",
       "Failure to Supply - Weighted Avg Exposure (2019)           float64\n",
       "Forecast                                                   float64\n",
       "Launch Date                                         datetime64[ns]\n",
       "Launch Date Serial                                         float64\n",
       "Supply Quantity (Replenishment)                            float64\n",
       "Market Share %                                             float64\n",
       "Number of Competitors                                      float64\n",
       "WAC Price                                                  float64\n",
       "FTS_indicator                                                int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create FTS_indicator to help model identify when a FTS rate is available or not \n",
    "bid_data_df['FTS_indicator'] = bid_data_df.apply(lambda row: 0 if row['Failure to Supply Rate (FTS) per unit'] == 'not available' else 1, axis=1)\n",
    "\n",
    "# Convert NaN's to 0's \n",
    "bid_data_df['Failure to Supply Rate (FTS) per unit'] = [0 if isinstance(x, str) else float(x) for x in list(bid_data_df['Failure to Supply Rate (FTS) per unit'].values)]\n",
    "bid_data_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                                float64\n",
       "Material Number                                            float64\n",
       "Product Family                                             float64\n",
       "Annual units                                               float64\n",
       "Monthly units                                              float64\n",
       "Contract Status                                             object\n",
       "Customer Supply Start date                                 float64\n",
       "Bid Submitted Date                                         float64\n",
       "Bid Submitted Month                                        float64\n",
       "Bid Submitted Year                                         float64\n",
       "Apotex Response                                             object\n",
       "Customer Response                                           object\n",
       "Failure to Supply Rate (FTS) per unit                      float64\n",
       "Failure to Supply - Weighted Avg Exposure (2018)           float64\n",
       "Failure to Supply - Weighted Avg Exposure (2019)           float64\n",
       "Forecast                                                   float64\n",
       "Launch Date                                         datetime64[ns]\n",
       "Launch Date Serial                                         float64\n",
       "Supply Quantity (Replenishment)                            float64\n",
       "Market Share %                                             float64\n",
       "Number of Competitors                                      float64\n",
       "WAC Price                                                  float64\n",
       "FTS_indicator                                                int64\n",
       "FTS_WAE_2018_indicator                                       int64\n",
       "FTS_WAE_2019_indicator                                       int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create FTS_WAE_2018_indicator to help model identify when a FTS WAE is available or not \n",
    "bid_data_df['FTS_WAE_2018_indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Failure to Supply - Weighted Avg Exposure (2018)']) == True else 1, axis=1)\n",
    "bid_data_df['Failure to Supply - Weighted Avg Exposure (2018)'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Failure to Supply - Weighted Avg Exposure (2018)'].values)]\n",
    "\n",
    "# Create FTS_WAE_2019_indicator to help model identify when a FTS WAE is available or not \n",
    "bid_data_df['FTS_WAE_2019_indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Failure to Supply - Weighted Avg Exposure (2019)']) == True else 1, axis=1)\n",
    "bid_data_df['Failure to Supply - Weighted Avg Exposure (2019)'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Failure to Supply - Weighted Avg Exposure (2019)'].values)]\n",
    "\n",
    "bid_data_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                                float64\n",
       "Material Number                                            float64\n",
       "Product Family                                             float64\n",
       "Annual units                                               float64\n",
       "Monthly units                                              float64\n",
       "Contract Status                                             object\n",
       "Customer Supply Start date                                 float64\n",
       "Bid Submitted Date                                         float64\n",
       "Bid Submitted Month                                        float64\n",
       "Bid Submitted Year                                         float64\n",
       "Apotex Response                                             object\n",
       "Customer Response                                           object\n",
       "Failure to Supply Rate (FTS) per unit                      float64\n",
       "Failure to Supply - Weighted Avg Exposure (2018)           float64\n",
       "Failure to Supply - Weighted Avg Exposure (2019)           float64\n",
       "Forecast                                                   float64\n",
       "Launch Date                                         datetime64[ns]\n",
       "Launch Date Serial                                         float64\n",
       "Supply Quantity (Replenishment)                            float64\n",
       "Market Share %                                             float64\n",
       "Number of Competitors                                      float64\n",
       "WAC Price                                                  float64\n",
       "FTS_indicator                                                int64\n",
       "FTS_WAE_2018_indicator                                       int64\n",
       "FTS_WAE_2019_indicator                                       int64\n",
       "Supply Quantity_Indicator                                    int64\n",
       "Number of Competitors_Indicator                              int64\n",
       "WAC Price_Indicator                                          int64\n",
       "Market Share % Indicator                                     int64\n",
       "Forecast_Indicator                                           int64\n",
       "Launch Date Serial_Indicator                                 int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create additional indicators for the model to identify when Supply Quantity, # of Competitors, WAC Price is available or not \n",
    "bid_data_df['Supply Quantity_Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Supply Quantity (Replenishment)']) == True else 1, axis=1)\n",
    "bid_data_df['Supply Quantity (Replenishment)'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Supply Quantity (Replenishment)'].values)]\n",
    "\n",
    "bid_data_df['Number of Competitors_Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Number of Competitors']) == True else 1, axis=1)\n",
    "bid_data_df['Number of Competitors'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Number of Competitors'].values)]\n",
    "\n",
    "bid_data_df['WAC Price_Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['WAC Price']) == True else 1, axis=1)\n",
    "bid_data_df['WAC Price'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['WAC Price'].values)]\n",
    "\n",
    "bid_data_df['Market Share % Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Market Share %']) == True else 1, axis=1)\n",
    "bid_data_df['Market Share %'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Market Share %'].values)]\n",
    "\n",
    "bid_data_df['Forecast_Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Forecast']) == True else 1, axis=1)\n",
    "bid_data_df['Forecast'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Forecast'].values)]\n",
    "\n",
    "bid_data_df['Launch Date Serial_Indicator'] = bid_data_df.apply(lambda row: 0 if math.isnan(row['Launch Date Serial']) == True else 1, axis=1)\n",
    "bid_data_df['Launch Date Serial'] = [0 if math.isnan(x)==True else float(x) for x in list(bid_data_df['Launch Date Serial'].values)]\n",
    "\n",
    "bid_data_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Material Number</th>\n",
       "      <th>Product Family</th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Month</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Apotex Response</th>\n",
       "      <th>...</th>\n",
       "      <th>FTS_WAE_2018_indicator</th>\n",
       "      <th>FTS_WAE_2019_indicator</th>\n",
       "      <th>Supply Quantity_Indicator</th>\n",
       "      <th>Number of Competitors_Indicator</th>\n",
       "      <th>WAC Price_Indicator</th>\n",
       "      <th>Market Share % Indicator</th>\n",
       "      <th>Forecast_Indicator</th>\n",
       "      <th>Launch Date Serial_Indicator</th>\n",
       "      <th>Contract Status_EXISTING BUSINESS</th>\n",
       "      <th>Contract Status_NEW BUSINESS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51025.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Approved</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51024.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Approved</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48912.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Approved</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48914.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Approved</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1000813.0</td>\n",
       "      <td>59481.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Approved</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer ID  Material Number  Product Family  Annual units  Monthly units  \\\n",
       "0    1001517.0          51025.0           381.0       65617.0    5468.000000   \n",
       "1    1001517.0          51024.0           381.0       54765.0    4564.000000   \n",
       "2    1001517.0          48912.0           381.0      121861.0   10155.000000   \n",
       "3    1001517.0          48914.0           381.0      101706.0    8476.000000   \n",
       "4    1000813.0          59481.0           322.0        4190.0     349.166667   \n",
       "\n",
       "   Customer Supply Start date  Bid Submitted Date  Bid Submitted Month  \\\n",
       "0                1.484024e+09        1.483229e+09                  1.0   \n",
       "1                1.484024e+09        1.483229e+09                  1.0   \n",
       "2                1.484024e+09        1.483229e+09                  1.0   \n",
       "3                1.484024e+09        1.483229e+09                  1.0   \n",
       "4                1.486184e+09        1.483229e+09                  1.0   \n",
       "\n",
       "   Bid Submitted Year Apotex Response  ... FTS_WAE_2018_indicator  \\\n",
       "0              2017.0        Approved  ...                      0   \n",
       "1              2017.0        Approved  ...                      0   \n",
       "2              2017.0        Approved  ...                      0   \n",
       "3              2017.0        Approved  ...                      0   \n",
       "4              2017.0        Approved  ...                      1   \n",
       "\n",
       "   FTS_WAE_2019_indicator  Supply Quantity_Indicator  \\\n",
       "0                       0                          0   \n",
       "1                       0                          0   \n",
       "2                       0                          0   \n",
       "3                       0                          0   \n",
       "4                       1                          1   \n",
       "\n",
       "   Number of Competitors_Indicator  WAC Price_Indicator  \\\n",
       "0                                0                    0   \n",
       "1                                0                    0   \n",
       "2                                0                    0   \n",
       "3                                0                    0   \n",
       "4                                0                    0   \n",
       "\n",
       "  Market Share % Indicator  Forecast_Indicator  Launch Date Serial_Indicator  \\\n",
       "0                        0                   1                             1   \n",
       "1                        0                   1                             1   \n",
       "2                        0                   1                             1   \n",
       "3                        0                   1                             1   \n",
       "4                        0                   1                             1   \n",
       "\n",
       "   Contract Status_EXISTING BUSINESS  Contract Status_NEW BUSINESS  \n",
       "0                                  0                             1  \n",
       "1                                  0                             1  \n",
       "2                                  0                             1  \n",
       "3                                  0                             1  \n",
       "4                                  0                             1  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# One hot encoding the Contract Status\n",
    "bid_data_df['Contract Status'] = pd.Categorical(bid_data_df['Contract Status'])\n",
    "dfDummies = pd.get_dummies(bid_data_df['Contract Status'], prefix = 'Contract Status')\n",
    "bid_data_df = bid_data_df.drop(['Contract Status'],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                                float64\n",
       "Material Number                                            float64\n",
       "Product Family                                             float64\n",
       "Annual units                                               float64\n",
       "Monthly units                                              float64\n",
       "Customer Supply Start date                                 float64\n",
       "Bid Submitted Date                                         float64\n",
       "Bid Submitted Year                                         float64\n",
       "Apotex Response                                             object\n",
       "Customer Response                                           object\n",
       "Failure to Supply Rate (FTS) per unit                      float64\n",
       "Failure to Supply - Weighted Avg Exposure (2018)           float64\n",
       "Failure to Supply - Weighted Avg Exposure (2019)           float64\n",
       "Forecast                                                   float64\n",
       "Launch Date                                         datetime64[ns]\n",
       "Launch Date Serial                                         float64\n",
       "Supply Quantity (Replenishment)                            float64\n",
       "Market Share %                                             float64\n",
       "Number of Competitors                                      float64\n",
       "WAC Price                                                  float64\n",
       "FTS_indicator                                                int64\n",
       "FTS_WAE_2018_indicator                                       int64\n",
       "FTS_WAE_2019_indicator                                       int64\n",
       "Supply Quantity_Indicator                                    int64\n",
       "Number of Competitors_Indicator                              int64\n",
       "WAC Price_Indicator                                          int64\n",
       "Market Share % Indicator                                     int64\n",
       "Forecast_Indicator                                           int64\n",
       "Launch Date Serial_Indicator                                 int64\n",
       "Contract Status_EXISTING BUSINESS                            uint8\n",
       "Contract Status_NEW BUSINESS                                 uint8\n",
       "Bid Submitted Month_1.0                                      uint8\n",
       "Bid Submitted Month_2.0                                      uint8\n",
       "Bid Submitted Month_3.0                                      uint8\n",
       "Bid Submitted Month_4.0                                      uint8\n",
       "Bid Submitted Month_5.0                                      uint8\n",
       "Bid Submitted Month_6.0                                      uint8\n",
       "Bid Submitted Month_7.0                                      uint8\n",
       "Bid Submitted Month_8.0                                      uint8\n",
       "Bid Submitted Month_9.0                                      uint8\n",
       "Bid Submitted Month_10.0                                     uint8\n",
       "Bid Submitted Month_11.0                                     uint8\n",
       "Bid Submitted Month_12.0                                     uint8\n",
       "dtype: object"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# One hot encoding the Bid Submitted Month\n",
    "bid_data_df[\"Bid Submitted Month\"] = pd.Categorical(bid_data_df[\"Bid Submitted Month\"])\n",
    "dfDummies = pd.get_dummies(bid_data_df[\"Bid Submitted Month\"], prefix = \"Bid Submitted Month\")\n",
    "bid_data_df = bid_data_df.drop([\"Bid Submitted Month\"],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Material Number</th>\n",
       "      <th>Product Family</th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Customer Response</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>...</th>\n",
       "      <th>Bid Submitted Month_7.0</th>\n",
       "      <th>Bid Submitted Month_8.0</th>\n",
       "      <th>Bid Submitted Month_9.0</th>\n",
       "      <th>Bid Submitted Month_10.0</th>\n",
       "      <th>Bid Submitted Month_11.0</th>\n",
       "      <th>Bid Submitted Month_12.0</th>\n",
       "      <th>Apotex Response_Approved</th>\n",
       "      <th>Apotex Response_Approved with alternate supply date</th>\n",
       "      <th>Apotex Response_Reject- Apotex Cancellation</th>\n",
       "      <th>Apotex Response_Reject- Supply Constraint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51025.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Won</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>51024.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Won</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48912.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Won</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>48914.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Won</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1000813.0</td>\n",
       "      <td>59481.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>Won</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer ID  Material Number  Product Family  Annual units  Monthly units  \\\n",
       "0    1001517.0          51025.0           381.0       65617.0    5468.000000   \n",
       "1    1001517.0          51024.0           381.0       54765.0    4564.000000   \n",
       "2    1001517.0          48912.0           381.0      121861.0   10155.000000   \n",
       "3    1001517.0          48914.0           381.0      101706.0    8476.000000   \n",
       "4    1000813.0          59481.0           322.0        4190.0     349.166667   \n",
       "\n",
       "   Customer Supply Start date  Bid Submitted Date  Bid Submitted Year  \\\n",
       "0                1.484024e+09        1.483229e+09              2017.0   \n",
       "1                1.484024e+09        1.483229e+09              2017.0   \n",
       "2                1.484024e+09        1.483229e+09              2017.0   \n",
       "3                1.484024e+09        1.483229e+09              2017.0   \n",
       "4                1.486184e+09        1.483229e+09              2017.0   \n",
       "\n",
       "  Customer Response  Failure to Supply Rate (FTS) per unit  ...  \\\n",
       "0               Won                                    0.0  ...   \n",
       "1               Won                                    0.0  ...   \n",
       "2               Won                                    0.0  ...   \n",
       "3               Won                                    0.0  ...   \n",
       "4               Won                                    0.0  ...   \n",
       "\n",
       "   Bid Submitted Month_7.0  Bid Submitted Month_8.0  Bid Submitted Month_9.0  \\\n",
       "0                        0                        0                        0   \n",
       "1                        0                        0                        0   \n",
       "2                        0                        0                        0   \n",
       "3                        0                        0                        0   \n",
       "4                        0                        0                        0   \n",
       "\n",
       "  Bid Submitted Month_10.0  Bid Submitted Month_11.0  \\\n",
       "0                        0                         0   \n",
       "1                        0                         0   \n",
       "2                        0                         0   \n",
       "3                        0                         0   \n",
       "4                        0                         0   \n",
       "\n",
       "   Bid Submitted Month_12.0  Apotex Response_Approved  \\\n",
       "0                         0                         1   \n",
       "1                         0                         1   \n",
       "2                         0                         1   \n",
       "3                         0                         1   \n",
       "4                         0                         1   \n",
       "\n",
       "   Apotex Response_Approved with alternate supply date  \\\n",
       "0                                                  0     \n",
       "1                                                  0     \n",
       "2                                                  0     \n",
       "3                                                  0     \n",
       "4                                                  0     \n",
       "\n",
       "   Apotex Response_Reject- Apotex Cancellation  \\\n",
       "0                                            0   \n",
       "1                                            0   \n",
       "2                                            0   \n",
       "3                                            0   \n",
       "4                                            0   \n",
       "\n",
       "   Apotex Response_Reject- Supply Constraint  \n",
       "0                                          0  \n",
       "1                                          0  \n",
       "2                                          0  \n",
       "3                                          0  \n",
       "4                                          0  \n",
       "\n",
       "[5 rows x 46 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# One hot encoding the Apotex Response\n",
    "bid_data_df[\"Apotex Response\"] = pd.Categorical(bid_data_df[\"Apotex Response\"])\n",
    "dfDummies = pd.get_dummies(bid_data_df[\"Apotex Response\"], prefix = \"Apotex Response\")\n",
    "bid_data_df = bid_data_df.drop([\"Apotex Response\"],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                               0\n",
       "Material Number                                           0\n",
       "Product Family                                            0\n",
       "Annual units                                              0\n",
       "Monthly units                                             0\n",
       "Customer Supply Start date                                0\n",
       "Bid Submitted Date                                        0\n",
       "Bid Submitted Year                                        0\n",
       "Customer Response                                         0\n",
       "Failure to Supply Rate (FTS) per unit                     0\n",
       "Failure to Supply - Weighted Avg Exposure (2018)          0\n",
       "Failure to Supply - Weighted Avg Exposure (2019)          0\n",
       "Forecast                                                  0\n",
       "Launch Date                                            1618\n",
       "Launch Date Serial                                        0\n",
       "Supply Quantity (Replenishment)                           0\n",
       "Market Share %                                            0\n",
       "Number of Competitors                                     0\n",
       "WAC Price                                                 0\n",
       "FTS_indicator                                             0\n",
       "FTS_WAE_2018_indicator                                    0\n",
       "FTS_WAE_2019_indicator                                    0\n",
       "Supply Quantity_Indicator                                 0\n",
       "Number of Competitors_Indicator                           0\n",
       "WAC Price_Indicator                                       0\n",
       "Market Share % Indicator                                  0\n",
       "Forecast_Indicator                                        0\n",
       "Launch Date Serial_Indicator                              0\n",
       "Contract Status_EXISTING BUSINESS                         0\n",
       "Contract Status_NEW BUSINESS                              0\n",
       "Bid Submitted Month_1.0                                   0\n",
       "Bid Submitted Month_2.0                                   0\n",
       "Bid Submitted Month_3.0                                   0\n",
       "Bid Submitted Month_4.0                                   0\n",
       "Bid Submitted Month_5.0                                   0\n",
       "Bid Submitted Month_6.0                                   0\n",
       "Bid Submitted Month_7.0                                   0\n",
       "Bid Submitted Month_8.0                                   0\n",
       "Bid Submitted Month_9.0                                   0\n",
       "Bid Submitted Month_10.0                                  0\n",
       "Bid Submitted Month_11.0                                  0\n",
       "Bid Submitted Month_12.0                                  0\n",
       "Apotex Response_Approved                                  0\n",
       "Apotex Response_Approved with alternate supply date       0\n",
       "Apotex Response_Reject- Apotex Cancellation               0\n",
       "Apotex Response_Reject- Supply Constraint                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Count the NaN's in each column\n",
    "bid_data_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "bid_data_df = bid_data_df.drop(columns = ['Launch Date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Lost ', 'Won'}"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Address target variable anomalies (the Did not Bid entries) \n",
    "bid_data_df = bid_data_df[bid_data_df['Customer Response'] != 'Did not Bid- Apotex Cancellation']\n",
    "bid_data_df = bid_data_df[bid_data_df['Customer Response'] != 'Did not Bid- Supply']\n",
    "\n",
    "set(bid_data_df['Customer Response'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replacing Customer Response: Lost with 0 and Won with 1\n",
    "bid_data_df['Customer Response'] = bid_data_df.apply(\n",
    "    lambda row: 0 if row['Customer Response'] == 'Lost ' else 1,\n",
    "    axis=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Customer ID                                            0\n",
       "Material Number                                        0\n",
       "Product Family                                         0\n",
       "Annual units                                           0\n",
       "Monthly units                                          0\n",
       "Customer Supply Start date                             0\n",
       "Bid Submitted Date                                     0\n",
       "Bid Submitted Year                                     0\n",
       "Customer Response                                      0\n",
       "Failure to Supply Rate (FTS) per unit                  0\n",
       "Failure to Supply - Weighted Avg Exposure (2018)       0\n",
       "Failure to Supply - Weighted Avg Exposure (2019)       0\n",
       "Forecast                                               0\n",
       "Launch Date Serial                                     0\n",
       "Supply Quantity (Replenishment)                        0\n",
       "Market Share %                                         0\n",
       "Number of Competitors                                  0\n",
       "WAC Price                                              0\n",
       "FTS_indicator                                          0\n",
       "FTS_WAE_2018_indicator                                 0\n",
       "FTS_WAE_2019_indicator                                 0\n",
       "Supply Quantity_Indicator                              0\n",
       "Number of Competitors_Indicator                        0\n",
       "WAC Price_Indicator                                    0\n",
       "Market Share % Indicator                               0\n",
       "Forecast_Indicator                                     0\n",
       "Launch Date Serial_Indicator                           0\n",
       "Contract Status_EXISTING BUSINESS                      0\n",
       "Contract Status_NEW BUSINESS                           0\n",
       "Bid Submitted Month_1.0                                0\n",
       "Bid Submitted Month_2.0                                0\n",
       "Bid Submitted Month_3.0                                0\n",
       "Bid Submitted Month_4.0                                0\n",
       "Bid Submitted Month_5.0                                0\n",
       "Bid Submitted Month_6.0                                0\n",
       "Bid Submitted Month_7.0                                0\n",
       "Bid Submitted Month_8.0                                0\n",
       "Bid Submitted Month_9.0                                0\n",
       "Bid Submitted Month_10.0                               0\n",
       "Bid Submitted Month_11.0                               0\n",
       "Bid Submitted Month_12.0                               0\n",
       "Apotex Response_Approved                               0\n",
       "Apotex Response_Approved with alternate supply date    0\n",
       "Apotex Response_Reject- Apotex Cancellation            0\n",
       "Apotex Response_Reject- Supply Constraint              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Count the NaN's in each column\n",
    "bid_data_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Product Family</th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Customer Response</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2018)</th>\n",
       "      <th>...</th>\n",
       "      <th>Material Number_68127.0</th>\n",
       "      <th>Material Number_68128.0</th>\n",
       "      <th>Material Number_68129.0</th>\n",
       "      <th>Material Number_68181.0</th>\n",
       "      <th>Material Number_68185.0</th>\n",
       "      <th>Material Number_68186.0</th>\n",
       "      <th>Material Number_68187.0</th>\n",
       "      <th>Material Number_68347.0</th>\n",
       "      <th>Material Number_68358.0</th>\n",
       "      <th>Material Number_68359.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>381.0</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1000813.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.569672</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 571 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer ID  Product Family  Annual units  Monthly units  \\\n",
       "0    1001517.0           381.0       65617.0    5468.000000   \n",
       "1    1001517.0           381.0       54765.0    4564.000000   \n",
       "2    1001517.0           381.0      121861.0   10155.000000   \n",
       "3    1001517.0           381.0      101706.0    8476.000000   \n",
       "4    1000813.0           322.0        4190.0     349.166667   \n",
       "\n",
       "   Customer Supply Start date  Bid Submitted Date  Bid Submitted Year  \\\n",
       "0                1.484024e+09        1.483229e+09              2017.0   \n",
       "1                1.484024e+09        1.483229e+09              2017.0   \n",
       "2                1.484024e+09        1.483229e+09              2017.0   \n",
       "3                1.484024e+09        1.483229e+09              2017.0   \n",
       "4                1.486184e+09        1.483229e+09              2017.0   \n",
       "\n",
       "   Customer Response  Failure to Supply Rate (FTS) per unit  \\\n",
       "0                  1                                    0.0   \n",
       "1                  1                                    0.0   \n",
       "2                  1                                    0.0   \n",
       "3                  1                                    0.0   \n",
       "4                  1                                    0.0   \n",
       "\n",
       "   Failure to Supply - Weighted Avg Exposure (2018)  ...  \\\n",
       "0                                          0.000000  ...   \n",
       "1                                          0.000000  ...   \n",
       "2                                          0.000000  ...   \n",
       "3                                          0.000000  ...   \n",
       "4                                          2.569672  ...   \n",
       "\n",
       "   Material Number_68127.0  Material Number_68128.0  Material Number_68129.0  \\\n",
       "0                        0                        0                        0   \n",
       "1                        0                        0                        0   \n",
       "2                        0                        0                        0   \n",
       "3                        0                        0                        0   \n",
       "4                        0                        0                        0   \n",
       "\n",
       "   Material Number_68181.0  Material Number_68185.0  Material Number_68186.0  \\\n",
       "0                        0                        0                        0   \n",
       "1                        0                        0                        0   \n",
       "2                        0                        0                        0   \n",
       "3                        0                        0                        0   \n",
       "4                        0                        0                        0   \n",
       "\n",
       "   Material Number_68187.0  Material Number_68347.0  Material Number_68358.0  \\\n",
       "0                        0                        0                        0   \n",
       "1                        0                        0                        0   \n",
       "2                        0                        0                        0   \n",
       "3                        0                        0                        0   \n",
       "4                        0                        0                        0   \n",
       "\n",
       "   Material Number_68359.0  \n",
       "0                        0  \n",
       "1                        0  \n",
       "2                        0  \n",
       "3                        0  \n",
       "4                        0  \n",
       "\n",
       "[5 rows x 571 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now that our data types are set, we should one-hot encode the Material Numbers (better model prediction) \n",
    "bid_data_df[\"Material Number\"] = pd.Categorical(bid_data_df[\"Material Number\"])\n",
    "dfDummies = pd.get_dummies(bid_data_df[\"Material Number\"], prefix = \"Material Number\")\n",
    "bid_data_df = bid_data_df.drop([\"Material Number\"],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bid_data_df[\"Product Family\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Customer Response</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2018)</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2019)</th>\n",
       "      <th>...</th>\n",
       "      <th>Product Family_996.0</th>\n",
       "      <th>Product Family_998.0</th>\n",
       "      <th>Product Family_1007.0</th>\n",
       "      <th>Product Family_1009.0</th>\n",
       "      <th>Product Family_1057.0</th>\n",
       "      <th>Product Family_1064.0</th>\n",
       "      <th>Product Family_1068.0</th>\n",
       "      <th>Product Family_1119.0</th>\n",
       "      <th>Product Family_1149.0</th>\n",
       "      <th>Product Family_1155.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1001517.0</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1000813.0</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.569672</td>\n",
       "      <td>0.225163</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 683 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer ID  Annual units  Monthly units  Customer Supply Start date  \\\n",
       "0    1001517.0       65617.0    5468.000000                1.484024e+09   \n",
       "1    1001517.0       54765.0    4564.000000                1.484024e+09   \n",
       "2    1001517.0      121861.0   10155.000000                1.484024e+09   \n",
       "3    1001517.0      101706.0    8476.000000                1.484024e+09   \n",
       "4    1000813.0        4190.0     349.166667                1.486184e+09   \n",
       "\n",
       "   Bid Submitted Date  Bid Submitted Year  Customer Response  \\\n",
       "0        1.483229e+09              2017.0                  1   \n",
       "1        1.483229e+09              2017.0                  1   \n",
       "2        1.483229e+09              2017.0                  1   \n",
       "3        1.483229e+09              2017.0                  1   \n",
       "4        1.483229e+09              2017.0                  1   \n",
       "\n",
       "   Failure to Supply Rate (FTS) per unit  \\\n",
       "0                                    0.0   \n",
       "1                                    0.0   \n",
       "2                                    0.0   \n",
       "3                                    0.0   \n",
       "4                                    0.0   \n",
       "\n",
       "   Failure to Supply - Weighted Avg Exposure (2018)  \\\n",
       "0                                          0.000000   \n",
       "1                                          0.000000   \n",
       "2                                          0.000000   \n",
       "3                                          0.000000   \n",
       "4                                          2.569672   \n",
       "\n",
       "   Failure to Supply - Weighted Avg Exposure (2019)  ...  \\\n",
       "0                                          0.000000  ...   \n",
       "1                                          0.000000  ...   \n",
       "2                                          0.000000  ...   \n",
       "3                                          0.000000  ...   \n",
       "4                                          0.225163  ...   \n",
       "\n",
       "   Product Family_996.0  Product Family_998.0  Product Family_1007.0  \\\n",
       "0                     0                     0                      0   \n",
       "1                     0                     0                      0   \n",
       "2                     0                     0                      0   \n",
       "3                     0                     0                      0   \n",
       "4                     0                     0                      0   \n",
       "\n",
       "   Product Family_1009.0  Product Family_1057.0  Product Family_1064.0  \\\n",
       "0                      0                      0                      0   \n",
       "1                      0                      0                      0   \n",
       "2                      0                      0                      0   \n",
       "3                      0                      0                      0   \n",
       "4                      0                      0                      0   \n",
       "\n",
       "   Product Family_1068.0  Product Family_1119.0  Product Family_1149.0  \\\n",
       "0                      0                      0                      0   \n",
       "1                      0                      0                      0   \n",
       "2                      0                      0                      0   \n",
       "3                      0                      0                      0   \n",
       "4                      0                      0                      0   \n",
       "\n",
       "   Product Family_1155.0  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  \n",
       "\n",
       "[5 rows x 683 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We should also one-hot encode the Product Family \n",
    "bid_data_df[\"Product Family\"] = pd.Categorical(bid_data_df[\"Product Family\"])\n",
    "dfDummies = pd.get_dummies(bid_data_df[\"Product Family\"], prefix = \"Product Family\")\n",
    "bid_data_df = bid_data_df.drop([\"Product Family\"],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Customer Response</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2018)</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2019)</th>\n",
       "      <th>Forecast</th>\n",
       "      <th>...</th>\n",
       "      <th>Customer ID_1002554.0</th>\n",
       "      <th>Customer ID_1002568.0</th>\n",
       "      <th>Customer ID_1002588.0</th>\n",
       "      <th>Customer ID_1002638.0</th>\n",
       "      <th>Customer ID_1002663.0</th>\n",
       "      <th>Customer ID_1002670.0</th>\n",
       "      <th>Customer ID_1002713.0</th>\n",
       "      <th>Customer ID_1002736.0</th>\n",
       "      <th>Customer ID_1002741.0</th>\n",
       "      <th>Customer ID_1002743.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>905.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1043.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2008.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.569672</td>\n",
       "      <td>0.225163</td>\n",
       "      <td>522.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Annual units  Monthly units  Customer Supply Start date  \\\n",
       "0       65617.0    5468.000000                1.484024e+09   \n",
       "1       54765.0    4564.000000                1.484024e+09   \n",
       "2      121861.0   10155.000000                1.484024e+09   \n",
       "3      101706.0    8476.000000                1.484024e+09   \n",
       "4        4190.0     349.166667                1.486184e+09   \n",
       "\n",
       "   Bid Submitted Date  Bid Submitted Year  Customer Response  \\\n",
       "0        1.483229e+09              2017.0                  1   \n",
       "1        1.483229e+09              2017.0                  1   \n",
       "2        1.483229e+09              2017.0                  1   \n",
       "3        1.483229e+09              2017.0                  1   \n",
       "4        1.483229e+09              2017.0                  1   \n",
       "\n",
       "   Failure to Supply Rate (FTS) per unit  \\\n",
       "0                                    0.0   \n",
       "1                                    0.0   \n",
       "2                                    0.0   \n",
       "3                                    0.0   \n",
       "4                                    0.0   \n",
       "\n",
       "   Failure to Supply - Weighted Avg Exposure (2018)  \\\n",
       "0                                          0.000000   \n",
       "1                                          0.000000   \n",
       "2                                          0.000000   \n",
       "3                                          0.000000   \n",
       "4                                          2.569672   \n",
       "\n",
       "   Failure to Supply - Weighted Avg Exposure (2019)  Forecast  ...  \\\n",
       "0                                          0.000000     905.0  ...   \n",
       "1                                          0.000000    1400.0  ...   \n",
       "2                                          0.000000    1043.0  ...   \n",
       "3                                          0.000000    2008.0  ...   \n",
       "4                                          0.225163     522.0  ...   \n",
       "\n",
       "   Customer ID_1002554.0  Customer ID_1002568.0  Customer ID_1002588.0  \\\n",
       "0                      0                      0                      0   \n",
       "1                      0                      0                      0   \n",
       "2                      0                      0                      0   \n",
       "3                      0                      0                      0   \n",
       "4                      0                      0                      0   \n",
       "\n",
       "   Customer ID_1002638.0  Customer ID_1002663.0  Customer ID_1002670.0  \\\n",
       "0                      0                      0                      0   \n",
       "1                      0                      0                      0   \n",
       "2                      0                      0                      0   \n",
       "3                      0                      0                      0   \n",
       "4                      0                      0                      0   \n",
       "\n",
       "   Customer ID_1002713.0  Customer ID_1002736.0  Customer ID_1002741.0  \\\n",
       "0                      0                      0                      0   \n",
       "1                      0                      0                      0   \n",
       "2                      0                      0                      0   \n",
       "3                      0                      0                      0   \n",
       "4                      0                      0                      0   \n",
       "\n",
       "   Customer ID_1002743.0  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  \n",
       "\n",
       "[5 rows x 784 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We should also one-hot encode the Customer IDs \n",
    "bid_data_df[\"Customer ID\"] = pd.Categorical(bid_data_df[\"Customer ID\"])\n",
    "dfDummies = pd.get_dummies(bid_data_df[\"Customer ID\"], prefix = \"Customer ID\")\n",
    "bid_data_df = bid_data_df.drop([\"Customer ID\"],axis=1)\n",
    "bid_data_df = pd.concat([bid_data_df, dfDummies], axis=1)\n",
    "bid_data_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Annual units</th>\n",
       "      <th>Monthly units</th>\n",
       "      <th>Customer Supply Start date</th>\n",
       "      <th>Bid Submitted Date</th>\n",
       "      <th>Bid Submitted Year</th>\n",
       "      <th>Customer Response</th>\n",
       "      <th>Failure to Supply Rate (FTS) per unit</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2018)</th>\n",
       "      <th>Failure to Supply - Weighted Avg Exposure (2019)</th>\n",
       "      <th>Forecast</th>\n",
       "      <th>...</th>\n",
       "      <th>Customer ID_1002554.0</th>\n",
       "      <th>Customer ID_1002568.0</th>\n",
       "      <th>Customer ID_1002588.0</th>\n",
       "      <th>Customer ID_1002638.0</th>\n",
       "      <th>Customer ID_1002663.0</th>\n",
       "      <th>Customer ID_1002670.0</th>\n",
       "      <th>Customer ID_1002713.0</th>\n",
       "      <th>Customer ID_1002736.0</th>\n",
       "      <th>Customer ID_1002741.0</th>\n",
       "      <th>Customer ID_1002743.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>65617.0</td>\n",
       "      <td>5468.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>905.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>54765.0</td>\n",
       "      <td>4564.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>121861.0</td>\n",
       "      <td>10155.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1043.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>101706.0</td>\n",
       "      <td>8476.000000</td>\n",
       "      <td>1.484024e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2008.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4190.0</td>\n",
       "      <td>349.166667</td>\n",
       "      <td>1.486184e+09</td>\n",
       "      <td>1.483229e+09</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.569672</td>\n",
       "      <td>0.225163</td>\n",
       "      <td>522.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12334</td>\n",
       "      <td>1152.0</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>1.576818e+09</td>\n",
       "      <td>1.575158e+09</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1</td>\n",
       "      <td>51.800040</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>51.800040</td>\n",
       "      <td>366.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12335</td>\n",
       "      <td>1800.0</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>1.576818e+09</td>\n",
       "      <td>1.575158e+09</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4.500937</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.500937</td>\n",
       "      <td>2483.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12336</td>\n",
       "      <td>288.0</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>1.576818e+09</td>\n",
       "      <td>1.575158e+09</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1</td>\n",
       "      <td>29.084799</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.084799</td>\n",
       "      <td>267.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12337</td>\n",
       "      <td>1152.0</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>1.576818e+09</td>\n",
       "      <td>1.575158e+09</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5.360276</td>\n",
       "      <td>2.926313</td>\n",
       "      <td>5.360276</td>\n",
       "      <td>1123.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12338</td>\n",
       "      <td>1200.0</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.576818e+09</td>\n",
       "      <td>1.575158e+09</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1</td>\n",
       "      <td>13.433112</td>\n",
       "      <td>11.369215</td>\n",
       "      <td>13.433112</td>\n",
       "      <td>270.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11385 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Annual units  Monthly units  Customer Supply Start date  \\\n",
       "0           65617.0    5468.000000                1.484024e+09   \n",
       "1           54765.0    4564.000000                1.484024e+09   \n",
       "2          121861.0   10155.000000                1.484024e+09   \n",
       "3          101706.0    8476.000000                1.484024e+09   \n",
       "4            4190.0     349.166667                1.486184e+09   \n",
       "...             ...            ...                         ...   \n",
       "12334        1152.0      96.000000                1.576818e+09   \n",
       "12335        1800.0     150.000000                1.576818e+09   \n",
       "12336         288.0      24.000000                1.576818e+09   \n",
       "12337        1152.0      96.000000                1.576818e+09   \n",
       "12338        1200.0     100.000000                1.576818e+09   \n",
       "\n",
       "       Bid Submitted Date  Bid Submitted Year  Customer Response  \\\n",
       "0            1.483229e+09              2017.0                  1   \n",
       "1            1.483229e+09              2017.0                  1   \n",
       "2            1.483229e+09              2017.0                  1   \n",
       "3            1.483229e+09              2017.0                  1   \n",
       "4            1.483229e+09              2017.0                  1   \n",
       "...                   ...                 ...                ...   \n",
       "12334        1.575158e+09              2019.0                  1   \n",
       "12335        1.575158e+09              2019.0                  1   \n",
       "12336        1.575158e+09              2019.0                  1   \n",
       "12337        1.575158e+09              2019.0                  1   \n",
       "12338        1.575158e+09              2019.0                  1   \n",
       "\n",
       "       Failure to Supply Rate (FTS) per unit  \\\n",
       "0                                   0.000000   \n",
       "1                                   0.000000   \n",
       "2                                   0.000000   \n",
       "3                                   0.000000   \n",
       "4                                   0.000000   \n",
       "...                                      ...   \n",
       "12334                              51.800040   \n",
       "12335                               4.500937   \n",
       "12336                              29.084799   \n",
       "12337                               5.360276   \n",
       "12338                              13.433112   \n",
       "\n",
       "       Failure to Supply - Weighted Avg Exposure (2018)  \\\n",
       "0                                              0.000000   \n",
       "1                                              0.000000   \n",
       "2                                              0.000000   \n",
       "3                                              0.000000   \n",
       "4                                              2.569672   \n",
       "...                                                 ...   \n",
       "12334                                          0.000000   \n",
       "12335                                          0.000000   \n",
       "12336                                          0.000000   \n",
       "12337                                          2.926313   \n",
       "12338                                         11.369215   \n",
       "\n",
       "       Failure to Supply - Weighted Avg Exposure (2019)  Forecast  ...  \\\n",
       "0                                              0.000000     905.0  ...   \n",
       "1                                              0.000000    1400.0  ...   \n",
       "2                                              0.000000    1043.0  ...   \n",
       "3                                              0.000000    2008.0  ...   \n",
       "4                                              0.225163     522.0  ...   \n",
       "...                                                 ...       ...  ...   \n",
       "12334                                         51.800040     366.0  ...   \n",
       "12335                                          4.500937    2483.0  ...   \n",
       "12336                                         29.084799     267.0  ...   \n",
       "12337                                          5.360276    1123.0  ...   \n",
       "12338                                         13.433112     270.0  ...   \n",
       "\n",
       "       Customer ID_1002554.0  Customer ID_1002568.0  Customer ID_1002588.0  \\\n",
       "0                          0                      0                      0   \n",
       "1                          0                      0                      0   \n",
       "2                          0                      0                      0   \n",
       "3                          0                      0                      0   \n",
       "4                          0                      0                      0   \n",
       "...                      ...                    ...                    ...   \n",
       "12334                      0                      0                      0   \n",
       "12335                      0                      0                      0   \n",
       "12336                      0                      0                      0   \n",
       "12337                      0                      0                      0   \n",
       "12338                      0                      0                      0   \n",
       "\n",
       "       Customer ID_1002638.0  Customer ID_1002663.0  Customer ID_1002670.0  \\\n",
       "0                          0                      0                      0   \n",
       "1                          0                      0                      0   \n",
       "2                          0                      0                      0   \n",
       "3                          0                      0                      0   \n",
       "4                          0                      0                      0   \n",
       "...                      ...                    ...                    ...   \n",
       "12334                      0                      0                      0   \n",
       "12335                      0                      0                      0   \n",
       "12336                      0                      0                      0   \n",
       "12337                      0                      0                      0   \n",
       "12338                      0                      0                      0   \n",
       "\n",
       "       Customer ID_1002713.0  Customer ID_1002736.0  Customer ID_1002741.0  \\\n",
       "0                          0                      0                      0   \n",
       "1                          0                      0                      0   \n",
       "2                          0                      0                      0   \n",
       "3                          0                      0                      0   \n",
       "4                          0                      0                      0   \n",
       "...                      ...                    ...                    ...   \n",
       "12334                      0                      0                      0   \n",
       "12335                      0                      0                      0   \n",
       "12336                      0                      0                      0   \n",
       "12337                      0                      0                      0   \n",
       "12338                      0                      0                      0   \n",
       "\n",
       "       Customer ID_1002743.0  \n",
       "0                          0  \n",
       "1                          0  \n",
       "2                          0  \n",
       "3                          0  \n",
       "4                          0  \n",
       "...                      ...  \n",
       "12334                      0  \n",
       "12335                      0  \n",
       "12336                      0  \n",
       "12337                      0  \n",
       "12338                      0  \n",
       "\n",
       "[11385 rows x 784 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bid_data_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Annual units                                           float64\n",
      "Monthly units                                          float64\n",
      "Customer Supply Start date                             float64\n",
      "Bid Submitted Date                                     float64\n",
      "Bid Submitted Year                                     float64\n",
      "Customer Response                                        int64\n",
      "Failure to Supply Rate (FTS) per unit                  float64\n",
      "Failure to Supply - Weighted Avg Exposure (2018)       float64\n",
      "Failure to Supply - Weighted Avg Exposure (2019)       float64\n",
      "Forecast                                               float64\n",
      "Launch Date Serial                                     float64\n",
      "Supply Quantity (Replenishment)                        float64\n",
      "Market Share %                                         float64\n",
      "Number of Competitors                                  float64\n",
      "WAC Price                                              float64\n",
      "FTS_indicator                                            int64\n",
      "FTS_WAE_2018_indicator                                   int64\n",
      "FTS_WAE_2019_indicator                                   int64\n",
      "Supply Quantity_Indicator                                int64\n",
      "Number of Competitors_Indicator                          int64\n",
      "WAC Price_Indicator                                      int64\n",
      "Market Share % Indicator                                 int64\n",
      "Forecast_Indicator                                       int64\n",
      "Launch Date Serial_Indicator                             int64\n",
      "Contract Status_EXISTING BUSINESS                        uint8\n",
      "Contract Status_NEW BUSINESS                             uint8\n",
      "Bid Submitted Month_1.0                                  uint8\n",
      "Bid Submitted Month_2.0                                  uint8\n",
      "Bid Submitted Month_3.0                                  uint8\n",
      "Bid Submitted Month_4.0                                  uint8\n",
      "Bid Submitted Month_5.0                                  uint8\n",
      "Bid Submitted Month_6.0                                  uint8\n",
      "Bid Submitted Month_7.0                                  uint8\n",
      "Bid Submitted Month_8.0                                  uint8\n",
      "Bid Submitted Month_9.0                                  uint8\n",
      "Bid Submitted Month_10.0                                 uint8\n",
      "Bid Submitted Month_11.0                                 uint8\n",
      "Bid Submitted Month_12.0                                 uint8\n",
      "Apotex Response_Approved                                 uint8\n",
      "Apotex Response_Approved with alternate supply date      uint8\n",
      "Apotex Response_Reject- Apotex Cancellation              uint8\n",
      "Apotex Response_Reject- Supply Constraint                uint8\n",
      "Material Number_42374.0                                  uint8\n",
      "Material Number_42376.0                                  uint8\n",
      "Material Number_42401.0                                  uint8\n",
      "Material Number_42403.0                                  uint8\n",
      "Material Number_42405.0                                  uint8\n",
      "Material Number_42407.0                                  uint8\n",
      "Material Number_42409.0                                  uint8\n",
      "Material Number_42410.0                                  uint8\n",
      "Material Number_42418.0                                  uint8\n",
      "Material Number_42420.0                                  uint8\n",
      "Material Number_42422.0                                  uint8\n",
      "Material Number_42424.0                                  uint8\n",
      "Material Number_42433.0                                  uint8\n",
      "Material Number_42435.0                                  uint8\n",
      "Material Number_42443.0                                  uint8\n",
      "Material Number_42445.0                                  uint8\n",
      "Material Number_42447.0                                  uint8\n",
      "Material Number_42457.0                                  uint8\n",
      "Material Number_42458.0                                  uint8\n",
      "Material Number_42459.0                                  uint8\n",
      "Material Number_42460.0                                  uint8\n",
      "Material Number_42461.0                                  uint8\n",
      "Material Number_42462.0                                  uint8\n",
      "Material Number_42463.0                                  uint8\n",
      "Material Number_42464.0                                  uint8\n",
      "Material Number_42465.0                                  uint8\n",
      "Material Number_42466.0                                  uint8\n",
      "Material Number_42467.0                                  uint8\n",
      "Material Number_42468.0                                  uint8\n",
      "Material Number_42469.0                                  uint8\n",
      "Material Number_42470.0                                  uint8\n",
      "Material Number_42482.0                                  uint8\n",
      "Material Number_42483.0                                  uint8\n",
      "Material Number_42484.0                                  uint8\n",
      "Material Number_42485.0                                  uint8\n",
      "Material Number_42516.0                                  uint8\n",
      "Material Number_42517.0                                  uint8\n",
      "Material Number_42518.0                                  uint8\n",
      "Material Number_42561.0                                  uint8\n",
      "Material Number_42562.0                                  uint8\n",
      "Material Number_42563.0                                  uint8\n",
      "Material Number_42564.0                                  uint8\n",
      "Material Number_42565.0                                  uint8\n",
      "Material Number_42566.0                                  uint8\n",
      "Material Number_42567.0                                  uint8\n",
      "Material Number_42569.0                                  uint8\n",
      "Material Number_42571.0                                  uint8\n",
      "Material Number_42573.0                                  uint8\n",
      "Material Number_42575.0                                  uint8\n",
      "Material Number_42577.0                                  uint8\n",
      "Material Number_42579.0                                  uint8\n",
      "Material Number_42581.0                                  uint8\n",
      "Material Number_42582.0                                  uint8\n",
      "Material Number_42583.0                                  uint8\n",
      "Material Number_42585.0                                  uint8\n",
      "Material Number_42586.0                                  uint8\n",
      "Material Number_42588.0                                  uint8\n",
      "Material Number_42589.0                                  uint8\n",
      "Material Number_42591.0                                  uint8\n",
      "Material Number_42613.0                                  uint8\n",
      "Material Number_42615.0                                  uint8\n",
      "Material Number_42620.0                                  uint8\n",
      "Material Number_42623.0                                  uint8\n",
      "Material Number_42624.0                                  uint8\n",
      "Material Number_42625.0                                  uint8\n",
      "Material Number_42646.0                                  uint8\n",
      "Material Number_42647.0                                  uint8\n",
      "Material Number_42648.0                                  uint8\n",
      "Material Number_42649.0                                  uint8\n",
      "Material Number_42654.0                                  uint8\n",
      "Material Number_42655.0                                  uint8\n",
      "Material Number_42657.0                                  uint8\n",
      "Material Number_42658.0                                  uint8\n",
      "Material Number_42660.0                                  uint8\n",
      "Material Number_42661.0                                  uint8\n",
      "Material Number_43658.0                                  uint8\n",
      "Material Number_43986.0                                  uint8\n",
      "Material Number_44035.0                                  uint8\n",
      "Material Number_44038.0                                  uint8\n",
      "Material Number_44042.0                                  uint8\n",
      "Material Number_44050.0                                  uint8\n",
      "Material Number_44052.0                                  uint8\n",
      "Material Number_44054.0                                  uint8\n",
      "Material Number_44058.0                                  uint8\n",
      "Material Number_44059.0                                  uint8\n",
      "Material Number_44060.0                                  uint8\n",
      "Material Number_44582.0                                  uint8\n",
      "Material Number_44584.0                                  uint8\n",
      "Material Number_44968.0                                  uint8\n",
      "Material Number_44970.0                                  uint8\n",
      "Material Number_44980.0                                  uint8\n",
      "Material Number_44981.0                                  uint8\n",
      "Material Number_45316.0                                  uint8\n",
      "Material Number_45320.0                                  uint8\n",
      "Material Number_45392.0                                  uint8\n",
      "Material Number_45393.0                                  uint8\n",
      "Material Number_45429.0                                  uint8\n",
      "Material Number_45430.0                                  uint8\n",
      "Material Number_45531.0                                  uint8\n",
      "Material Number_45534.0                                  uint8\n",
      "Material Number_45545.0                                  uint8\n",
      "Material Number_45709.0                                  uint8\n",
      "Material Number_45720.0                                  uint8\n",
      "Material Number_45726.0                                  uint8\n",
      "Material Number_45968.0                                  uint8\n",
      "Material Number_46094.0                                  uint8\n",
      "Material Number_46108.0                                  uint8\n",
      "Material Number_46119.0                                  uint8\n",
      "Material Number_46125.0                                  uint8\n",
      "Material Number_46126.0                                  uint8\n",
      "Material Number_46128.0                                  uint8\n",
      "Material Number_46256.0                                  uint8\n",
      "Material Number_46308.0                                  uint8\n",
      "Material Number_46384.0                                  uint8\n",
      "Material Number_46385.0                                  uint8\n",
      "Material Number_46386.0                                  uint8\n",
      "Material Number_46387.0                                  uint8\n",
      "Material Number_46408.0                                  uint8\n",
      "Material Number_46409.0                                  uint8\n",
      "Material Number_46612.0                                  uint8\n",
      "Material Number_46677.0                                  uint8\n",
      "Material Number_46714.0                                  uint8\n",
      "Material Number_46716.0                                  uint8\n",
      "Material Number_46717.0                                  uint8\n",
      "Material Number_46808.0                                  uint8\n",
      "Material Number_46814.0                                  uint8\n",
      "Material Number_46815.0                                  uint8\n",
      "Material Number_46816.0                                  uint8\n",
      "Material Number_46834.0                                  uint8\n",
      "Material Number_46922.0                                  uint8\n",
      "Material Number_46929.0                                  uint8\n",
      "Material Number_46972.0                                  uint8\n",
      "Material Number_46973.0                                  uint8\n",
      "Material Number_46974.0                                  uint8\n",
      "Material Number_46981.0                                  uint8\n",
      "Material Number_46988.0                                  uint8\n",
      "Material Number_46989.0                                  uint8\n",
      "Material Number_47059.0                                  uint8\n",
      "Material Number_47065.0                                  uint8\n",
      "Material Number_47135.0                                  uint8\n",
      "Material Number_47138.0                                  uint8\n",
      "Material Number_47149.0                                  uint8\n",
      "Material Number_47151.0                                  uint8\n",
      "Material Number_47154.0                                  uint8\n",
      "Material Number_47155.0                                  uint8\n",
      "Material Number_47157.0                                  uint8\n",
      "Material Number_47161.0                                  uint8\n",
      "Material Number_47166.0                                  uint8\n",
      "Material Number_47251.0                                  uint8\n",
      "Material Number_47254.0                                  uint8\n",
      "Material Number_47257.0                                  uint8\n",
      "Material Number_47713.0                                  uint8\n",
      "Material Number_47715.0                                  uint8\n",
      "Material Number_47717.0                                  uint8\n",
      "Material Number_47719.0                                  uint8\n",
      "Material Number_47723.0                                  uint8\n",
      "Material Number_47805.0                                  uint8\n",
      "Material Number_47808.0                                  uint8\n",
      "Material Number_47852.0                                  uint8\n",
      "Material Number_47854.0                                  uint8\n",
      "Material Number_47986.0                                  uint8\n",
      "Material Number_48009.0                                  uint8\n",
      "Material Number_48695.0                                  uint8\n",
      "Material Number_48696.0                                  uint8\n",
      "Material Number_48697.0                                  uint8\n",
      "Material Number_48698.0                                  uint8\n",
      "Material Number_48740.0                                  uint8\n",
      "Material Number_48912.0                                  uint8\n",
      "Material Number_48913.0                                  uint8\n",
      "Material Number_48914.0                                  uint8\n",
      "Material Number_48916.0                                  uint8\n",
      "Material Number_49048.0                                  uint8\n",
      "Material Number_49086.0                                  uint8\n",
      "Material Number_49087.0                                  uint8\n",
      "Material Number_49088.0                                  uint8\n",
      "Material Number_49092.0                                  uint8\n",
      "Material Number_49216.0                                  uint8\n",
      "Material Number_49288.0                                  uint8\n",
      "Material Number_49560.0                                  uint8\n",
      "Material Number_49572.0                                  uint8\n",
      "Material Number_50574.0                                  uint8\n",
      "Material Number_50589.0                                  uint8\n",
      "Material Number_50595.0                                  uint8\n",
      "Material Number_50848.0                                  uint8\n",
      "Material Number_50998.0                                  uint8\n",
      "Material Number_51012.0                                  uint8\n",
      "Material Number_51024.0                                  uint8\n",
      "Material Number_51025.0                                  uint8\n",
      "Material Number_51304.0                                  uint8\n",
      "Material Number_51305.0                                  uint8\n",
      "Material Number_51306.0                                  uint8\n",
      "Material Number_51307.0                                  uint8\n",
      "Material Number_51317.0                                  uint8\n",
      "Material Number_51328.0                                  uint8\n",
      "Material Number_51329.0                                  uint8\n",
      "Material Number_51330.0                                  uint8\n",
      "Material Number_51331.0                                  uint8\n",
      "Material Number_51332.0                                  uint8\n",
      "Material Number_51333.0                                  uint8\n",
      "Material Number_51451.0                                  uint8\n",
      "Material Number_51452.0                                  uint8\n",
      "Material Number_51453.0                                  uint8\n",
      "Material Number_51523.0                                  uint8\n",
      "Material Number_51552.0                                  uint8\n",
      "Material Number_51553.0                                  uint8\n",
      "Material Number_51795.0                                  uint8\n",
      "Material Number_51796.0                                  uint8\n",
      "Material Number_52029.0                                  uint8\n",
      "Material Number_52035.0                                  uint8\n",
      "Material Number_52054.0                                  uint8\n",
      "Material Number_52055.0                                  uint8\n",
      "Material Number_52057.0                                  uint8\n",
      "Material Number_52592.0                                  uint8\n",
      "Material Number_52593.0                                  uint8\n",
      "Material Number_52594.0                                  uint8\n",
      "Material Number_52610.0                                  uint8\n",
      "Material Number_52611.0                                  uint8\n",
      "Material Number_52612.0                                  uint8\n",
      "Material Number_52613.0                                  uint8\n",
      "Material Number_52614.0                                  uint8\n",
      "Material Number_52727.0                                  uint8\n",
      "Material Number_52998.0                                  uint8\n",
      "Material Number_53003.0                                  uint8\n",
      "Material Number_53080.0                                  uint8\n",
      "Material Number_53086.0                                  uint8\n",
      "Material Number_53102.0                                  uint8\n",
      "Material Number_53108.0                                  uint8\n",
      "Material Number_53114.0                                  uint8\n",
      "Material Number_53120.0                                  uint8\n",
      "Material Number_53126.0                                  uint8\n",
      "Material Number_53145.0                                  uint8\n",
      "Material Number_53148.0                                  uint8\n",
      "Material Number_53151.0                                  uint8\n",
      "Material Number_53212.0                                  uint8\n",
      "Material Number_53970.0                                  uint8\n",
      "Material Number_54050.0                                  uint8\n",
      "Material Number_54069.0                                  uint8\n",
      "Material Number_54071.0                                  uint8\n",
      "Material Number_54073.0                                  uint8\n",
      "Material Number_54077.0                                  uint8\n",
      "Material Number_54101.0                                  uint8\n",
      "Material Number_54107.0                                  uint8\n",
      "Material Number_54112.0                                  uint8\n",
      "Material Number_54121.0                                  uint8\n",
      "Material Number_54123.0                                  uint8\n",
      "Material Number_54127.0                                  uint8\n",
      "Material Number_54129.0                                  uint8\n",
      "Material Number_54134.0                                  uint8\n",
      "Material Number_54136.0                                  uint8\n",
      "Material Number_54141.0                                  uint8\n",
      "Material Number_54143.0                                  uint8\n",
      "Material Number_54145.0                                  uint8\n",
      "Material Number_54146.0                                  uint8\n",
      "Material Number_54147.0                                  uint8\n",
      "Material Number_54170.0                                  uint8\n",
      "Material Number_54880.0                                  uint8\n",
      "Material Number_55624.0                                  uint8\n",
      "Material Number_55784.0                                  uint8\n",
      "Material Number_55972.0                                  uint8\n",
      "Material Number_55984.0                                  uint8\n",
      "Material Number_56008.0                                  uint8\n",
      "Material Number_56041.0                                  uint8\n",
      "Material Number_56042.0                                  uint8\n",
      "Material Number_56043.0                                  uint8\n",
      "Material Number_56044.0                                  uint8\n",
      "Material Number_56628.0                                  uint8\n",
      "Material Number_56784.0                                  uint8\n",
      "Material Number_57338.0                                  uint8\n",
      "Material Number_57552.0                                  uint8\n",
      "Material Number_57553.0                                  uint8\n",
      "Material Number_57554.0                                  uint8\n",
      "Material Number_57787.0                                  uint8\n",
      "Material Number_57975.0                                  uint8\n",
      "Material Number_58366.0                                  uint8\n",
      "Material Number_58367.0                                  uint8\n",
      "Material Number_58557.0                                  uint8\n",
      "Material Number_58578.0                                  uint8\n",
      "Material Number_58580.0                                  uint8\n",
      "Material Number_58581.0                                  uint8\n",
      "Material Number_58582.0                                  uint8\n",
      "Material Number_58583.0                                  uint8\n",
      "Material Number_58584.0                                  uint8\n",
      "Material Number_58585.0                                  uint8\n",
      "Material Number_58678.0                                  uint8\n",
      "Material Number_58679.0                                  uint8\n",
      "Material Number_58680.0                                  uint8\n",
      "Material Number_58681.0                                  uint8\n",
      "Material Number_58682.0                                  uint8\n",
      "Material Number_58683.0                                  uint8\n",
      "Material Number_58806.0                                  uint8\n",
      "Material Number_58807.0                                  uint8\n",
      "Material Number_58878.0                                  uint8\n",
      "Material Number_58879.0                                  uint8\n",
      "Material Number_59245.0                                  uint8\n",
      "Material Number_59428.0                                  uint8\n",
      "Material Number_59465.0                                  uint8\n",
      "Material Number_59466.0                                  uint8\n",
      "Material Number_59467.0                                  uint8\n",
      "Material Number_59474.0                                  uint8\n",
      "Material Number_59478.0                                  uint8\n",
      "Material Number_59480.0                                  uint8\n",
      "Material Number_59481.0                                  uint8\n",
      "Material Number_59483.0                                  uint8\n",
      "Material Number_59484.0                                  uint8\n",
      "Material Number_59486.0                                  uint8\n",
      "Material Number_59487.0                                  uint8\n",
      "Material Number_59489.0                                  uint8\n",
      "Material Number_59490.0                                  uint8\n",
      "Material Number_59492.0                                  uint8\n",
      "Material Number_59493.0                                  uint8\n",
      "Material Number_59495.0                                  uint8\n",
      "Material Number_59499.0                                  uint8\n",
      "Material Number_59509.0                                  uint8\n",
      "Material Number_60119.0                                  uint8\n",
      "Material Number_60157.0                                  uint8\n",
      "Material Number_60288.0                                  uint8\n",
      "Material Number_60335.0                                  uint8\n",
      "Material Number_60336.0                                  uint8\n",
      "Material Number_60337.0                                  uint8\n",
      "Material Number_60394.0                                  uint8\n",
      "Material Number_60648.0                                  uint8\n",
      "Material Number_60649.0                                  uint8\n",
      "Material Number_60650.0                                  uint8\n",
      "Material Number_60651.0                                  uint8\n",
      "Material Number_60652.0                                  uint8\n",
      "Material Number_60653.0                                  uint8\n",
      "Material Number_60654.0                                  uint8\n",
      "Material Number_60655.0                                  uint8\n",
      "Material Number_60820.0                                  uint8\n",
      "Material Number_61084.0                                  uint8\n",
      "Material Number_61086.0                                  uint8\n",
      "Material Number_61100.0                                  uint8\n",
      "Material Number_61217.0                                  uint8\n",
      "Material Number_61441.0                                  uint8\n",
      "Material Number_61457.0                                  uint8\n",
      "Material Number_61464.0                                  uint8\n",
      "Material Number_61555.0                                  uint8\n",
      "Material Number_61557.0                                  uint8\n",
      "Material Number_61568.0                                  uint8\n",
      "Material Number_61569.0                                  uint8\n",
      "Material Number_61572.0                                  uint8\n",
      "Material Number_61574.0                                  uint8\n",
      "Material Number_61586.0                                  uint8\n",
      "Material Number_61587.0                                  uint8\n",
      "Material Number_61589.0                                  uint8\n",
      "Material Number_61590.0                                  uint8\n",
      "Material Number_61592.0                                  uint8\n",
      "Material Number_62073.0                                  uint8\n",
      "Material Number_62074.0                                  uint8\n",
      "Material Number_62253.0                                  uint8\n",
      "Material Number_62728.0                                  uint8\n",
      "Material Number_62767.0                                  uint8\n",
      "Material Number_63022.0                                  uint8\n",
      "Material Number_63023.0                                  uint8\n",
      "Material Number_63024.0                                  uint8\n",
      "Material Number_63025.0                                  uint8\n",
      "Material Number_63026.0                                  uint8\n",
      "Material Number_63207.0                                  uint8\n",
      "Material Number_63299.0                                  uint8\n",
      "Material Number_63300.0                                  uint8\n",
      "Material Number_63307.0                                  uint8\n",
      "Material Number_63318.0                                  uint8\n",
      "Material Number_63441.0                                  uint8\n",
      "Material Number_63443.0                                  uint8\n",
      "Material Number_63444.0                                  uint8\n",
      "Material Number_63445.0                                  uint8\n",
      "Material Number_63479.0                                  uint8\n",
      "Material Number_63895.0                                  uint8\n",
      "Material Number_63896.0                                  uint8\n",
      "Material Number_63907.0                                  uint8\n",
      "Material Number_63918.0                                  uint8\n",
      "Material Number_63982.0                                  uint8\n",
      "Material Number_63984.0                                  uint8\n",
      "Material Number_64047.0                                  uint8\n",
      "Material Number_64049.0                                  uint8\n",
      "Material Number_64141.0                                  uint8\n",
      "Material Number_64142.0                                  uint8\n",
      "Material Number_64143.0                                  uint8\n",
      "Material Number_64271.0                                  uint8\n",
      "Material Number_64640.0                                  uint8\n",
      "Material Number_64799.0                                  uint8\n",
      "Material Number_65048.0                                  uint8\n",
      "Material Number_65050.0                                  uint8\n",
      "Material Number_65051.0                                  uint8\n",
      "Material Number_65052.0                                  uint8\n",
      "Material Number_65053.0                                  uint8\n",
      "Material Number_65054.0                                  uint8\n",
      "Material Number_65055.0                                  uint8\n",
      "Material Number_65056.0                                  uint8\n",
      "Material Number_65058.0                                  uint8\n",
      "Material Number_65059.0                                  uint8\n",
      "Material Number_65060.0                                  uint8\n",
      "Material Number_65062.0                                  uint8\n",
      "Material Number_65064.0                                  uint8\n",
      "Material Number_65065.0                                  uint8\n",
      "Material Number_65066.0                                  uint8\n",
      "Material Number_65159.0                                  uint8\n",
      "Material Number_65160.0                                  uint8\n",
      "Material Number_65161.0                                  uint8\n",
      "Material Number_65162.0                                  uint8\n",
      "Material Number_65242.0                                  uint8\n",
      "Material Number_65243.0                                  uint8\n",
      "Material Number_65568.0                                  uint8\n",
      "Material Number_65569.0                                  uint8\n",
      "Material Number_65570.0                                  uint8\n",
      "Material Number_65571.0                                  uint8\n",
      "Material Number_65759.0                                  uint8\n",
      "Material Number_65760.0                                  uint8\n",
      "Material Number_65765.0                                  uint8\n",
      "Material Number_65942.0                                  uint8\n",
      "Material Number_65943.0                                  uint8\n",
      "Material Number_65944.0                                  uint8\n",
      "Material Number_65945.0                                  uint8\n",
      "Material Number_65946.0                                  uint8\n",
      "Material Number_65959.0                                  uint8\n",
      "Material Number_65960.0                                  uint8\n",
      "Material Number_65961.0                                  uint8\n",
      "Material Number_65984.0                                  uint8\n",
      "Material Number_66143.0                                  uint8\n",
      "Material Number_66172.0                                  uint8\n",
      "Material Number_66345.0                                  uint8\n",
      "Material Number_66346.0                                  uint8\n",
      "Material Number_66401.0                                  uint8\n",
      "Material Number_66416.0                                  uint8\n",
      "Material Number_66492.0                                  uint8\n",
      "Material Number_66520.0                                  uint8\n",
      "Material Number_66589.0                                  uint8\n",
      "Material Number_66710.0                                  uint8\n",
      "Material Number_67415.0                                  uint8\n",
      "Material Number_67416.0                                  uint8\n",
      "Material Number_67417.0                                  uint8\n",
      "Material Number_67456.0                                  uint8\n",
      "Material Number_67458.0                                  uint8\n",
      "Material Number_67459.0                                  uint8\n",
      "Material Number_67460.0                                  uint8\n",
      "Material Number_67461.0                                  uint8\n",
      "Material Number_67467.0                                  uint8\n",
      "Material Number_67469.0                                  uint8\n",
      "Material Number_67471.0                                  uint8\n",
      "Material Number_67473.0                                  uint8\n",
      "Material Number_67474.0                                  uint8\n",
      "Material Number_67475.0                                  uint8\n",
      "Material Number_67476.0                                  uint8\n",
      "Material Number_67478.0                                  uint8\n",
      "Material Number_67504.0                                  uint8\n",
      "Material Number_67505.0                                  uint8\n",
      "Material Number_67525.0                                  uint8\n",
      "Material Number_67526.0                                  uint8\n",
      "Material Number_67527.0                                  uint8\n",
      "Material Number_67528.0                                  uint8\n",
      "Material Number_67529.0                                  uint8\n",
      "Material Number_67530.0                                  uint8\n",
      "Material Number_67531.0                                  uint8\n",
      "Material Number_67532.0                                  uint8\n",
      "Material Number_67533.0                                  uint8\n",
      "Material Number_67534.0                                  uint8\n",
      "Material Number_67551.0                                  uint8\n",
      "Material Number_67552.0                                  uint8\n",
      "Material Number_67553.0                                  uint8\n",
      "Material Number_67554.0                                  uint8\n",
      "Material Number_67555.0                                  uint8\n",
      "Material Number_67556.0                                  uint8\n",
      "Material Number_67557.0                                  uint8\n",
      "Material Number_67559.0                                  uint8\n",
      "Material Number_67578.0                                  uint8\n",
      "Material Number_67579.0                                  uint8\n",
      "Material Number_67580.0                                  uint8\n",
      "Material Number_67581.0                                  uint8\n",
      "Material Number_67584.0                                  uint8\n",
      "Material Number_67587.0                                  uint8\n",
      "Material Number_67628.0                                  uint8\n",
      "Material Number_67629.0                                  uint8\n",
      "Material Number_67630.0                                  uint8\n",
      "Material Number_67665.0                                  uint8\n",
      "Material Number_67666.0                                  uint8\n",
      "Material Number_67667.0                                  uint8\n",
      "Material Number_67722.0                                  uint8\n",
      "Material Number_67723.0                                  uint8\n",
      "Material Number_67724.0                                  uint8\n",
      "Material Number_67725.0                                  uint8\n",
      "Material Number_67726.0                                  uint8\n",
      "Material Number_67727.0                                  uint8\n",
      "Material Number_67768.0                                  uint8\n",
      "Material Number_67782.0                                  uint8\n",
      "Material Number_67783.0                                  uint8\n",
      "Material Number_67784.0                                  uint8\n",
      "Material Number_67785.0                                  uint8\n",
      "Material Number_67786.0                                  uint8\n",
      "Material Number_67787.0                                  uint8\n",
      "Material Number_67789.0                                  uint8\n",
      "Material Number_67793.0                                  uint8\n",
      "Material Number_67798.0                                  uint8\n",
      "Material Number_67799.0                                  uint8\n",
      "Material Number_67800.0                                  uint8\n",
      "Material Number_67801.0                                  uint8\n",
      "Material Number_67802.0                                  uint8\n",
      "Material Number_67953.0                                  uint8\n",
      "Material Number_67954.0                                  uint8\n",
      "Material Number_67955.0                                  uint8\n",
      "Material Number_67956.0                                  uint8\n",
      "Material Number_68101.0                                  uint8\n",
      "Material Number_68102.0                                  uint8\n",
      "Material Number_68103.0                                  uint8\n",
      "Material Number_68104.0                                  uint8\n",
      "Material Number_68105.0                                  uint8\n",
      "Material Number_68106.0                                  uint8\n",
      "Material Number_68109.0                                  uint8\n",
      "Material Number_68111.0                                  uint8\n",
      "Material Number_68112.0                                  uint8\n",
      "Material Number_68113.0                                  uint8\n",
      "Material Number_68114.0                                  uint8\n",
      "Material Number_68115.0                                  uint8\n",
      "Material Number_68116.0                                  uint8\n",
      "Material Number_68123.0                                  uint8\n",
      "Material Number_68124.0                                  uint8\n",
      "Material Number_68125.0                                  uint8\n",
      "Material Number_68126.0                                  uint8\n",
      "Material Number_68127.0                                  uint8\n",
      "Material Number_68128.0                                  uint8\n",
      "Material Number_68129.0                                  uint8\n",
      "Material Number_68181.0                                  uint8\n",
      "Material Number_68185.0                                  uint8\n",
      "Material Number_68186.0                                  uint8\n",
      "Material Number_68187.0                                  uint8\n",
      "Material Number_68347.0                                  uint8\n",
      "Material Number_68358.0                                  uint8\n",
      "Material Number_68359.0                                  uint8\n",
      "Product Family_4.0                                       uint8\n",
      "Product Family_28.0                                      uint8\n",
      "Product Family_29.0                                      uint8\n",
      "Product Family_34.0                                      uint8\n",
      "Product Family_52.0                                      uint8\n",
      "Product Family_53.0                                      uint8\n",
      "Product Family_57.0                                      uint8\n",
      "Product Family_59.0                                      uint8\n",
      "Product Family_61.0                                      uint8\n",
      "Product Family_67.0                                      uint8\n",
      "Product Family_69.0                                      uint8\n",
      "Product Family_72.0                                      uint8\n",
      "Product Family_74.0                                      uint8\n",
      "Product Family_83.0                                      uint8\n",
      "Product Family_87.0                                      uint8\n",
      "Product Family_90.0                                      uint8\n",
      "Product Family_107.0                                     uint8\n",
      "Product Family_117.0                                     uint8\n",
      "Product Family_127.0                                     uint8\n",
      "Product Family_152.0                                     uint8\n",
      "Product Family_153.0                                     uint8\n",
      "Product Family_159.0                                     uint8\n",
      "Product Family_161.0                                     uint8\n",
      "Product Family_166.0                                     uint8\n",
      "Product Family_167.0                                     uint8\n",
      "Product Family_179.0                                     uint8\n",
      "Product Family_182.0                                     uint8\n",
      "Product Family_201.0                                     uint8\n",
      "Product Family_203.0                                     uint8\n",
      "Product Family_217.0                                     uint8\n",
      "Product Family_218.0                                     uint8\n",
      "Product Family_228.0                                     uint8\n",
      "Product Family_229.0                                     uint8\n",
      "Product Family_232.0                                     uint8\n",
      "Product Family_234.0                                     uint8\n",
      "Product Family_238.0                                     uint8\n",
      "Product Family_240.0                                     uint8\n",
      "Product Family_258.0                                     uint8\n",
      "Product Family_269.0                                     uint8\n",
      "Product Family_272.0                                     uint8\n",
      "Product Family_274.0                                     uint8\n",
      "Product Family_287.0                                     uint8\n",
      "Product Family_293.0                                     uint8\n",
      "Product Family_295.0                                     uint8\n",
      "Product Family_309.0                                     uint8\n",
      "Product Family_311.0                                     uint8\n",
      "Product Family_313.0                                     uint8\n",
      "Product Family_318.0                                     uint8\n",
      "Product Family_320.0                                     uint8\n",
      "Product Family_322.0                                     uint8\n",
      "Product Family_329.0                                     uint8\n",
      "Product Family_330.0                                     uint8\n",
      "Product Family_333.0                                     uint8\n",
      "Product Family_335.0                                     uint8\n",
      "Product Family_348.0                                     uint8\n",
      "Product Family_353.0                                     uint8\n",
      "Product Family_354.0                                     uint8\n",
      "Product Family_355.0                                     uint8\n",
      "Product Family_358.0                                     uint8\n",
      "Product Family_365.0                                     uint8\n",
      "Product Family_366.0                                     uint8\n",
      "Product Family_368.0                                     uint8\n",
      "Product Family_371.0                                     uint8\n",
      "Product Family_373.0                                     uint8\n",
      "Product Family_374.0                                     uint8\n",
      "Product Family_376.0                                     uint8\n",
      "Product Family_381.0                                     uint8\n",
      "Product Family_391.0                                     uint8\n",
      "Product Family_394.0                                     uint8\n",
      "Product Family_404.0                                     uint8\n",
      "Product Family_405.0                                     uint8\n",
      "Product Family_415.0                                     uint8\n",
      "Product Family_424.0                                     uint8\n",
      "Product Family_427.0                                     uint8\n",
      "Product Family_434.0                                     uint8\n",
      "Product Family_441.0                                     uint8\n",
      "Product Family_455.0                                     uint8\n",
      "Product Family_460.0                                     uint8\n",
      "Product Family_465.0                                     uint8\n",
      "Product Family_468.0                                     uint8\n",
      "Product Family_469.0                                     uint8\n",
      "Product Family_470.0                                     uint8\n",
      "Product Family_475.0                                     uint8\n",
      "Product Family_478.0                                     uint8\n",
      "Product Family_492.0                                     uint8\n",
      "Product Family_497.0                                     uint8\n",
      "Product Family_507.0                                     uint8\n",
      "Product Family_512.0                                     uint8\n",
      "Product Family_513.0                                     uint8\n",
      "Product Family_518.0                                     uint8\n",
      "Product Family_520.0                                     uint8\n",
      "Product Family_536.0                                     uint8\n",
      "Product Family_545.0                                     uint8\n",
      "Product Family_547.0                                     uint8\n",
      "Product Family_553.0                                     uint8\n",
      "Product Family_562.0                                     uint8\n",
      "Product Family_574.0                                     uint8\n",
      "Product Family_594.0                                     uint8\n",
      "Product Family_597.0                                     uint8\n",
      "Product Family_625.0                                     uint8\n",
      "Product Family_638.0                                     uint8\n",
      "Product Family_673.0                                     uint8\n",
      "Product Family_714.0                                     uint8\n",
      "Product Family_996.0                                     uint8\n",
      "Product Family_998.0                                     uint8\n",
      "Product Family_1007.0                                    uint8\n",
      "Product Family_1009.0                                    uint8\n",
      "Product Family_1057.0                                    uint8\n",
      "Product Family_1064.0                                    uint8\n",
      "Product Family_1068.0                                    uint8\n",
      "Product Family_1119.0                                    uint8\n",
      "Product Family_1149.0                                    uint8\n",
      "Product Family_1155.0                                    uint8\n",
      "Customer ID_1000006.0                                    uint8\n",
      "Customer ID_1000010.0                                    uint8\n",
      "Customer ID_1000022.0                                    uint8\n",
      "Customer ID_1000042.0                                    uint8\n",
      "Customer ID_1000082.0                                    uint8\n",
      "Customer ID_1000094.0                                    uint8\n",
      "Customer ID_1000100.0                                    uint8\n",
      "Customer ID_1000108.0                                    uint8\n",
      "Customer ID_1000117.0                                    uint8\n",
      "Customer ID_1000136.0                                    uint8\n",
      "Customer ID_1000147.0                                    uint8\n",
      "Customer ID_1000169.0                                    uint8\n",
      "Customer ID_1000185.0                                    uint8\n",
      "Customer ID_1000197.0                                    uint8\n",
      "Customer ID_1000205.0                                    uint8\n",
      "Customer ID_1000222.0                                    uint8\n",
      "Customer ID_1000250.0                                    uint8\n",
      "Customer ID_1000256.0                                    uint8\n",
      "Customer ID_1000415.0                                    uint8\n",
      "Customer ID_1000485.0                                    uint8\n",
      "Customer ID_1000498.0                                    uint8\n",
      "Customer ID_1000545.0                                    uint8\n",
      "Customer ID_1000555.0                                    uint8\n",
      "Customer ID_1000587.0                                    uint8\n",
      "Customer ID_1000614.0                                    uint8\n",
      "Customer ID_1000620.0                                    uint8\n",
      "Customer ID_1000715.0                                    uint8\n",
      "Customer ID_1000720.0                                    uint8\n",
      "Customer ID_1000726.0                                    uint8\n",
      "Customer ID_1000748.0                                    uint8\n",
      "Customer ID_1000768.0                                    uint8\n",
      "Customer ID_1000804.0                                    uint8\n",
      "Customer ID_1000805.0                                    uint8\n",
      "Customer ID_1000806.0                                    uint8\n",
      "Customer ID_1000813.0                                    uint8\n",
      "Customer ID_1000824.0                                    uint8\n",
      "Customer ID_1000834.0                                    uint8\n",
      "Customer ID_1000858.0                                    uint8\n",
      "Customer ID_1000879.0                                    uint8\n",
      "Customer ID_1000908.0                                    uint8\n",
      "Customer ID_1000919.0                                    uint8\n",
      "Customer ID_1000939.0                                    uint8\n",
      "Customer ID_1000954.0                                    uint8\n",
      "Customer ID_1000993.0                                    uint8\n",
      "Customer ID_1001003.0                                    uint8\n",
      "Customer ID_1001008.0                                    uint8\n",
      "Customer ID_1001079.0                                    uint8\n",
      "Customer ID_1001089.0                                    uint8\n",
      "Customer ID_1001228.0                                    uint8\n",
      "Customer ID_1001270.0                                    uint8\n",
      "Customer ID_1001283.0                                    uint8\n",
      "Customer ID_1001328.0                                    uint8\n",
      "Customer ID_1001366.0                                    uint8\n",
      "Customer ID_1001383.0                                    uint8\n",
      "Customer ID_1001517.0                                    uint8\n",
      "Customer ID_1001539.0                                    uint8\n",
      "Customer ID_1001575.0                                    uint8\n",
      "Customer ID_1001856.0                                    uint8\n",
      "Customer ID_1001937.0                                    uint8\n",
      "Customer ID_1001960.0                                    uint8\n",
      "Customer ID_1001996.0                                    uint8\n",
      "Customer ID_1002018.0                                    uint8\n",
      "Customer ID_1002077.0                                    uint8\n",
      "Customer ID_1002109.0                                    uint8\n",
      "Customer ID_1002141.0                                    uint8\n",
      "Customer ID_1002143.0                                    uint8\n",
      "Customer ID_1002156.0                                    uint8\n",
      "Customer ID_1002168.0                                    uint8\n",
      "Customer ID_1002203.0                                    uint8\n",
      "Customer ID_1002211.0                                    uint8\n",
      "Customer ID_1002236.0                                    uint8\n",
      "Customer ID_1002244.0                                    uint8\n",
      "Customer ID_1002248.0                                    uint8\n",
      "Customer ID_1002258.0                                    uint8\n",
      "Customer ID_1002283.0                                    uint8\n",
      "Customer ID_1002291.0                                    uint8\n",
      "Customer ID_1002292.0                                    uint8\n",
      "Customer ID_1002341.0                                    uint8\n",
      "Customer ID_1002375.0                                    uint8\n",
      "Customer ID_1002413.0                                    uint8\n",
      "Customer ID_1002425.0                                    uint8\n",
      "Customer ID_1002446.0                                    uint8\n",
      "Customer ID_1002459.0                                    uint8\n",
      "Customer ID_1002484.0                                    uint8\n",
      "Customer ID_1002488.0                                    uint8\n",
      "Customer ID_1002496.0                                    uint8\n",
      "Customer ID_1002499.0                                    uint8\n",
      "Customer ID_1002513.0                                    uint8\n",
      "Customer ID_1002519.0                                    uint8\n",
      "Customer ID_1002520.0                                    uint8\n",
      "Customer ID_1002526.0                                    uint8\n",
      "Customer ID_1002550.0                                    uint8\n",
      "Customer ID_1002554.0                                    uint8\n",
      "Customer ID_1002568.0                                    uint8\n",
      "Customer ID_1002588.0                                    uint8\n",
      "Customer ID_1002638.0                                    uint8\n",
      "Customer ID_1002663.0                                    uint8\n",
      "Customer ID_1002670.0                                    uint8\n",
      "Customer ID_1002713.0                                    uint8\n",
      "Customer ID_1002736.0                                    uint8\n",
      "Customer ID_1002741.0                                    uint8\n",
      "Customer ID_1002743.0                                    uint8\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "with pd.option_context('display.max_rows', None, 'display.max_columns', None):\n",
    "    print(bid_data_df.dtypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "bid_data_df.to_csv('Bid_Data_Cleaned.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split dataset into target variable and feature variables \n",
    "df_X = bid_data_df.loc[:,bid_data_df.columns != 'Customer Response']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split dataset into target variable and feature variables \n",
    "df_y = bid_data_df.loc[:,[\"Customer Response\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split variables into Test-Train Split \n",
    "X_train, X_test, y_train, y_test = train_test_split(df_X, df_y, test_size=0.5, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                       n_jobs=None, oob_score=False, random_state=None,\n",
       "                       verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train RF classifier \n",
    "rf = RandomForestClassifier(n_estimators=100)\n",
    "rf.fit(X_train, y_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of RF classifier on test set: 0.81\n"
     ]
    }
   ],
   "source": [
    "# Predict with training set \n",
    "y_pred = rf.predict(X_test)\n",
    "print('Accuracy of RF classifier on test set: {:.2f}'.format(rf.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1756  673]\n",
      " [ 396 2868]]\n"
     ]
    }
   ],
   "source": [
    "# Confusion Matrix \n",
    "confusion_matrix = confusion_matrix(y_test, y_pred)\n",
    "print(confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.72      0.77      2429\n",
      "           1       0.81      0.88      0.84      3264\n",
      "\n",
      "    accuracy                           0.81      5693\n",
      "   macro avg       0.81      0.80      0.80      5693\n",
      "weighted avg       0.81      0.81      0.81      5693\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Classification Report \n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC \n",
    "rf_roc_auc = roc_auc_score(y_test, rf.predict(X_test))\n",
    "fpr, tpr, thresholds = roc_curve(y_test, rf.predict_proba(X_test)[:,1])\n",
    "plt.figure()\n",
    "plt.plot(fpr, tpr, label='RF (area = %0.2f)' % rf_roc_auc)\n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver operating characteristic')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.savefig('RF_ROC')\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "d = {'False Positive Rate': fpr, 'True Positive Rate': tpr}\n",
    "df = pd.DataFrame(d)\n",
    "sns.set(style='darkgrid')\n",
    "sns_plot = sns.lineplot(x='False Positive Rate', y='True Positive Rate', data=df).set_title(\"Receiver Operating Characteric (ROC)\")\n",
    "# sns_plot.text(0.85, 0.85,'Text Here', fontsize=9) \n",
    "sns_plot.figure.savefig(\"output.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Computing feature_importances \n",
    "feature_importances = pd.DataFrame(rf.feature_importances_,index = X_train.columns,\n",
    "                                   columns=['importance']).sort_values('importance',ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Customer ID</td>\n",
       "      <td>0.150940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Material Number</td>\n",
       "      <td>0.112003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Product Family</td>\n",
       "      <td>0.096329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Annual units</td>\n",
       "      <td>0.073983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Monthly units</td>\n",
       "      <td>0.073824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bid Submitted Month</td>\n",
       "      <td>0.071915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Customer Supply Start date</td>\n",
       "      <td>0.051064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Contract Status</td>\n",
       "      <td>0.042193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Forecast</td>\n",
       "      <td>0.039226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bid Submitted Date</td>\n",
       "      <td>0.037099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Failure to Supply - Weighted Avg Exposure (2018)</td>\n",
       "      <td>0.032780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Launch Date Serial</td>\n",
       "      <td>0.029202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Number of Competitors</td>\n",
       "      <td>0.026625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>WAC Price</td>\n",
       "      <td>0.025931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Failure to Supply - Weighted Avg Exposure (2019)</td>\n",
       "      <td>0.025135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Failure to Supply Rate (FTS) per unit</td>\n",
       "      <td>0.024282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Market Share %</td>\n",
       "      <td>0.022205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Supply Quantity (Replenishment)</td>\n",
       "      <td>0.016746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>FTS</td>\n",
       "      <td>0.015422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Apotex Response</td>\n",
       "      <td>0.012027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bid Submitted Year</td>\n",
       "      <td>0.010414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Supply Quantity</td>\n",
       "      <td>0.006807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Market Share % Indicator</td>\n",
       "      <td>0.003851</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  importance\n",
       "Customer ID                                         0.150940\n",
       "Material Number                                     0.112003\n",
       "Product Family                                      0.096329\n",
       "Annual units                                        0.073983\n",
       "Monthly units                                       0.073824\n",
       "Bid Submitted Month                                 0.071915\n",
       "Customer Supply Start date                          0.051064\n",
       "Contract Status                                     0.042193\n",
       "Forecast                                            0.039226\n",
       "Bid Submitted Date                                  0.037099\n",
       "Failure to Supply - Weighted Avg Exposure (2018)    0.032780\n",
       "Launch Date Serial                                  0.029202\n",
       "Number of Competitors                               0.026625\n",
       "WAC Price                                           0.025931\n",
       "Failure to Supply - Weighted Avg Exposure (2019)    0.025135\n",
       "Failure to Supply Rate (FTS) per unit               0.024282\n",
       "Market Share %                                      0.022205\n",
       "Supply Quantity (Replenishment)                     0.016746\n",
       "FTS                                                 0.015422\n",
       "Apotex Response                                     0.012027\n",
       "Bid Submitted Year                                  0.010414\n",
       "Supply Quantity                                     0.006807\n",
       "Market Share % Indicator                            0.003851"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_importances = feature_importances.groupby([s.split('_')[0] for s in feature_importances.index.values]).sum()\n",
    "feature_importances = feature_importances.sort_values('importance', ascending = False)\n",
    "feature_importances.to_excel('Feature_Importances.xlsx')\n",
    "feature_importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.6965305226174792, 0.6732542819499341, 0.7127799736495388, 0.7053140096618358, 0.6903820816864296]\n"
     ]
    }
   ],
   "source": [
    "# K-Fold cross-validation \n",
    "kf = KFold(n_splits=5) #KFold with k = 5\n",
    "scores = []\n",
    "for train_index, test_index in kf.split(df_X):\n",
    "    X_train, X_test = df_X.iloc[train_index], df_X.iloc[test_index]\n",
    "    y_train, y_test = df_y.iloc[train_index], df_y.iloc[test_index]\n",
    "    rf.fit(X_train, y_train.values.ravel())\n",
    "    scores.append(rf.score(X_test, y_test))\n",
    "print(scores) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of RF classifier with 5-fold Cross-Validation: 0.70\n"
     ]
    }
   ],
   "source": [
    "sum = 0\n",
    "avg = 0\n",
    "for i in scores:\n",
    "    sum += i \n",
    "avg = (sum/5.0) \n",
    "print('Accuracy of RF classifier with 5-fold Cross-Validation: {:.2f}'.format(avg))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.4 64-bit ('base': conda)",
   "language": "python",
   "name": "python37464bitbasecondaf8325423b44c4a6baa647795c44880d5"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}