model_development_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5fbf44e9",
   "metadata": {
    "colab_type": "text",
    "id": "view-in-github"
   },
   "source": [
    "<a href=\"https://colab.research.google.com/github/laurenemilyto/Project-4-Predicting-Hospial-Readmission-for-Diabetes/blob/main/diabetesv3.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8e5a08d",
   "metadata": {},
   "source": [
    "# Model Development"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "Fw-jWETP1-Sl",
   "metadata": {
    "id": "Fw-jWETP1-Sl"
   },
   "outputs": [],
   "source": [
    "#Loading libraries \n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import scipy as sp\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e75d6e4b",
   "metadata": {},
   "source": [
    "## Reading data from PostgresSQL Table\n",
    "table name: diabetes_db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "275e160b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import SQL Alchemy to read from Postgres databaase\n",
    "import psycopg2\n",
    "from sqlalchemy import create_engine, MetaData, func\n",
    "from sqlalchemy.ext.automap import automap_base\n",
    "from sqlalchemy.ext.declarative import declarative_base\n",
    "from sqlalchemy.orm import Session\n",
    "\n",
    "# Import modules to declare columns and column data types\n",
    "from sqlalchemy import Column, Integer, String, Float, Boolean, delete"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cpbN3g4nR0ny",
   "metadata": {
    "id": "cpbN3g4nR0ny"
   },
   "outputs": [],
   "source": [
    "# Establish connection to PostgreSQL\n",
    "# Modify password and localhost as needed\n",
    "engine = create_engine(\"postgresql://postgres:postgres@localhost:5432/diabetes_db\")\n",
    "conn = engine.connect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "62ca6eec",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read data from PostgreSQL database table and load into a DataFrame instance\n",
    "query_results = pd.read_sql(\"select * from \\\"diabetes\\\"\", conn);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0ee1f4c4",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          id             race  gender age time_in_hospital num_procedures  \\\n",
      "0          1        Caucasian  Female  15                3              0   \n",
      "1          2  AfricanAmerican  Female  25                2              5   \n",
      "2          3        Caucasian    Male  35                2              1   \n",
      "3          4        Caucasian    Male  45                1              0   \n",
      "4          5        Caucasian    Male  55                3              6   \n",
      "...      ...              ...     ...  ..              ...            ...   \n",
      "98047  98048  AfricanAmerican    Male  75                3              0   \n",
      "98048  98049  AfricanAmerican  Female  85                5              3   \n",
      "98049  98050        Caucasian    Male  75                1              0   \n",
      "98050  98051        Caucasian  Female  85               10              2   \n",
      "98051  98052        Caucasian    Male  75                6              3   \n",
      "\n",
      "      num_medications readmitted  \n",
      "0                  18          0  \n",
      "1                  13          0  \n",
      "2                  16          0  \n",
      "3                   8          0  \n",
      "4                  16          0  \n",
      "...               ...        ...  \n",
      "98047              16          0  \n",
      "98048              18          0  \n",
      "98049               9          0  \n",
      "98050              21          0  \n",
      "98051               3          0  \n",
      "\n",
      "[98052 rows x 8 columns]\n"
     ]
    }
   ],
   "source": [
    "# Print dataframe\n",
    "print(query_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecf4a79d",
   "metadata": {},
   "source": [
    "## Converting to Pandas Dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2094b13",
   "metadata": {},
   "source": [
    "- **Race**: Caucasian, Asian, African American, Hispanic and other\n",
    "- **Gender**: Male, Female\n",
    "- **Age**: Grouped in 10-year intervals [0, 10), [10, 20), ..., [90, 100)\n",
    "- **Time in Hospital**: Integer number of days between admission and discharge\n",
    "- **Number of Procedures**: Number of procedures (other than lab tests) performed during the encounter\n",
    "- **Number of medications**: Number of distinct generic names administered during the encounter\n",
    "- **Readmitted**: 'early readmission' is defined as within 30 days. Else 'other' if the patient was readmitted in more than 30 days or was not readmitted."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "EvG6rk0qVMGn",
   "metadata": {
    "id": "EvG6rk0qVMGn"
   },
   "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>id</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>readmitted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>55</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98047</th>\n",
       "      <td>98048</td>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98048</th>\n",
       "      <td>98049</td>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98049</th>\n",
       "      <td>98050</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98050</th>\n",
       "      <td>98051</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98051</th>\n",
       "      <td>98052</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98052 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id             race  gender age time_in_hospital num_procedures  \\\n",
       "0          1        Caucasian  Female  15                3              0   \n",
       "1          2  AfricanAmerican  Female  25                2              5   \n",
       "2          3        Caucasian    Male  35                2              1   \n",
       "3          4        Caucasian    Male  45                1              0   \n",
       "4          5        Caucasian    Male  55                3              6   \n",
       "...      ...              ...     ...  ..              ...            ...   \n",
       "98047  98048  AfricanAmerican    Male  75                3              0   \n",
       "98048  98049  AfricanAmerican  Female  85                5              3   \n",
       "98049  98050        Caucasian    Male  75                1              0   \n",
       "98050  98051        Caucasian  Female  85               10              2   \n",
       "98051  98052        Caucasian    Male  75                6              3   \n",
       "\n",
       "      num_medications readmitted  \n",
       "0                  18          0  \n",
       "1                  13          0  \n",
       "2                  16          0  \n",
       "3                   8          0  \n",
       "4                  16          0  \n",
       "...               ...        ...  \n",
       "98047              16          0  \n",
       "98048              18          0  \n",
       "98049               9          0  \n",
       "98050              21          0  \n",
       "98051               3          0  \n",
       "\n",
       "[98052 rows x 8 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(query_results)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed2dc543",
   "metadata": {},
   "source": [
    "## Preprocessing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2dfee9fc",
   "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>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>readmitted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>55</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98047</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98048</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98049</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98050</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98051</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98052 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  race  gender age time_in_hospital num_procedures  \\\n",
       "0            Caucasian  Female  15                3              0   \n",
       "1      AfricanAmerican  Female  25                2              5   \n",
       "2            Caucasian    Male  35                2              1   \n",
       "3            Caucasian    Male  45                1              0   \n",
       "4            Caucasian    Male  55                3              6   \n",
       "...                ...     ...  ..              ...            ...   \n",
       "98047  AfricanAmerican    Male  75                3              0   \n",
       "98048  AfricanAmerican  Female  85                5              3   \n",
       "98049        Caucasian    Male  75                1              0   \n",
       "98050        Caucasian  Female  85               10              2   \n",
       "98051        Caucasian    Male  75                6              3   \n",
       "\n",
       "      num_medications readmitted  \n",
       "0                  18          0  \n",
       "1                  13          0  \n",
       "2                  16          0  \n",
       "3                   8          0  \n",
       "4                  16          0  \n",
       "...               ...        ...  \n",
       "98047              16          0  \n",
       "98048              18          0  \n",
       "98049               9          0  \n",
       "98050              21          0  \n",
       "98051               3          0  \n",
       "\n",
       "[98052 rows x 7 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop ID column since this is not needed for our model\n",
    "df.drop(['id'], axis=1,inplace=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "HwlzGqJkIirQ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HwlzGqJkIirQ",
    "outputId": "ca5ff574-03a0-4006-86b1-951ab245442c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(98052, 7)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verify shape\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "785efc90",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "race                0\n",
       "gender              0\n",
       "age                 0\n",
       "time_in_hospital    0\n",
       "num_procedures      0\n",
       "num_medications     0\n",
       "readmitted          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verify that no null values exist\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1b3b141d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "race                object\n",
       "gender              object\n",
       "age                 object\n",
       "time_in_hospital    object\n",
       "num_procedures      object\n",
       "num_medications     object\n",
       "readmitted          object\n",
       "dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Data types currently = object\n",
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f9b27f98",
   "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>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>readmitted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>55</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98047</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98048</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98049</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98050</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>85</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98051</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>75</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98052 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  race  gender  age  time_in_hospital  num_procedures  \\\n",
       "0            Caucasian  Female   15                 3               0   \n",
       "1      AfricanAmerican  Female   25                 2               5   \n",
       "2            Caucasian    Male   35                 2               1   \n",
       "3            Caucasian    Male   45                 1               0   \n",
       "4            Caucasian    Male   55                 3               6   \n",
       "...                ...     ...  ...               ...             ...   \n",
       "98047  AfricanAmerican    Male   75                 3               0   \n",
       "98048  AfricanAmerican  Female   85                 5               3   \n",
       "98049        Caucasian    Male   75                 1               0   \n",
       "98050        Caucasian  Female   85                10               2   \n",
       "98051        Caucasian    Male   75                 6               3   \n",
       "\n",
       "       num_medications  readmitted  \n",
       "0                   18           0  \n",
       "1                   13           0  \n",
       "2                   16           0  \n",
       "3                    8           0  \n",
       "4                   16           0  \n",
       "...                ...         ...  \n",
       "98047               16           0  \n",
       "98048               18           0  \n",
       "98049                9           0  \n",
       "98050               21           0  \n",
       "98051                3           0  \n",
       "\n",
       "[98052 rows x 7 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert non-categorical data type to numeric \n",
    "df[['age', 'time_in_hospital', 'num_procedures', 'num_medications', \n",
    "    'readmitted']] = df[['age', 'time_in_hospital', 'num_procedures', 'num_medications', \n",
    "    'readmitted']].apply(pd.to_numeric)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "daeb5bb8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "race                object\n",
       "gender              object\n",
       "age                  int64\n",
       "time_in_hospital     int64\n",
       "num_procedures       int64\n",
       "num_medications      int64\n",
       "readmitted           int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verify data is converted to numeric\n",
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a02a6ac",
   "metadata": {},
   "source": [
    "### Encoding non-numerical columns: Label Encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "48bb428a",
   "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>race</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98047</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98048</th>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98049</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98050</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98051</th>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98052 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  race  gender\n",
       "0            Caucasian  Female\n",
       "1      AfricanAmerican  Female\n",
       "2            Caucasian    Male\n",
       "3            Caucasian    Male\n",
       "4            Caucasian    Male\n",
       "...                ...     ...\n",
       "98047  AfricanAmerican    Male\n",
       "98048  AfricanAmerican  Female\n",
       "98049        Caucasian    Male\n",
       "98050        Caucasian  Female\n",
       "98051        Caucasian    Male\n",
       "\n",
       "[98052 rows x 2 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_data = df.select_dtypes('O')\n",
    "num_data = df.select_dtypes(np.number)\n",
    "cat_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9da196ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'AfricanAmerican': 0, 'Asian': 1, 'Caucasian': 2, 'Hispanic': 3, 'Other': 4}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get categorical mapping for final application\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "le = LabelEncoder()\n",
    "le.fit(df['race'])\n",
    "integer_mapping = {l: i for i, l in enumerate(le.classes_)}\n",
    "integer_mapping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0661cdb8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Female': 0, 'Male': 1}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le = LabelEncoder()\n",
    "le.fit(df['gender'])\n",
    "integer_mapping = {l: i for i, l in enumerate(le.classes_)}\n",
    "integer_mapping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5aebe494",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-16-b2ef87209644>:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  cat_data[i] = le.fit_transform(cat_data[i])\n"
     ]
    }
   ],
   "source": [
    "# Encode categorical data\n",
    "le = LabelEncoder()\n",
    "for i in cat_data:\n",
    "  cat_data[i] = le.fit_transform(cat_data[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "4535e18c",
   "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>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>readmitted</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>55</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  time_in_hospital  num_procedures  num_medications  readmitted  race  \\\n",
       "0   15                 3               0               18           0     2   \n",
       "1   25                 2               5               13           0     0   \n",
       "2   35                 2               1               16           0     2   \n",
       "3   45                 1               0                8           0     2   \n",
       "4   55                 3               6               16           0     2   \n",
       "\n",
       "   gender  \n",
       "0       0  \n",
       "1       0  \n",
       "2       1  \n",
       "3       1  \n",
       "4       1  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.concat([num_data,cat_data],axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "5dc91c3b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(98052, 7)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3b8e8dd9",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"resources/final_data.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19865c0f",
   "metadata": {},
   "source": [
    "### Correlation Matrix\n",
    "**Understanding the relationship between attributes in our dataset.**\n",
    "\n",
    "Is there a correlation between features in our dataset? A strong positive correlation will be represented by a score of .9, and strong negative correlation will be represented by a score of -.9. If there is strong correlation between any features in our dataset, the model will be impacted by multicollinearity which can lead to misleading results. This will affect our Logistic Regression model. If features are strongly correlated, we will need to elimate them from our model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "8e4985e8",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Explore correlation between features via a heatmap\n",
    "fix, ax = plt.subplots(figsize = (18, 10))\n",
    "sns.set(font_scale = 2)\n",
    "sns.heatmap(df.corr(), annot=True, vmin=-1, vmax=1, center =0, cmap = 'coolwarm', linewidths=3, linecolor='black',)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e64113d",
   "metadata": {},
   "source": [
    "Looking at the correlation matrix, number of lab procedures and num_medications are slightly correlated (.38). Time in hospital and num_medications are also correlated (.46). However we do not believe that the relationship is strong enough to warrant feature removal. Since no features are strongly correlated in our matrix (above .5), we can move ahead with existing features. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17e3d409",
   "metadata": {},
   "source": [
    "### Defining Dependent & Independent Variables "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "2b752300",
   "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>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>55</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>16</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98047</th>\n",
       "      <td>75</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98048</th>\n",
       "      <td>85</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98049</th>\n",
       "      <td>75</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98050</th>\n",
       "      <td>85</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98051</th>\n",
       "      <td>75</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98052 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       age  time_in_hospital  num_procedures  num_medications  race  gender\n",
       "0       15                 3               0               18     2       0\n",
       "1       25                 2               5               13     0       0\n",
       "2       35                 2               1               16     2       1\n",
       "3       45                 1               0                8     2       1\n",
       "4       55                 3               6               16     2       1\n",
       "...    ...               ...             ...              ...   ...     ...\n",
       "98047   75                 3               0               16     0       1\n",
       "98048   85                 5               3               18     0       0\n",
       "98049   75                 1               0                9     2       1\n",
       "98050   85                10               2               21     2       0\n",
       "98051   75                 6               3                3     2       1\n",
       "\n",
       "[98052 rows x 6 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Split our preprocessed data into our features and target arrays\n",
    "y = df['readmitted']\n",
    "X = df.drop(columns = [\"readmitted\"], axis=1)\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e347226",
   "metadata": {},
   "source": [
    "### Verifying Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "IDDFPlKw7hpL",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IDDFPlKw7hpL",
    "outputId": "c5e8a5a9-a11a-4873-ba6c-6c813265a56b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of features in datast\n",
    "len(X.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "32e9b029",
   "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>age</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>num_procedures</th>\n",
       "      <th>num_medications</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>98052.000000</td>\n",
       "      <td>98052.000000</td>\n",
       "      <td>98052.000000</td>\n",
       "      <td>98052.000000</td>\n",
       "      <td>98052.000000</td>\n",
       "      <td>98052.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>66.260250</td>\n",
       "      <td>4.422011</td>\n",
       "      <td>1.350712</td>\n",
       "      <td>16.119590</td>\n",
       "      <td>1.658987</td>\n",
       "      <td>0.461174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>15.590805</td>\n",
       "      <td>2.993070</td>\n",
       "      <td>1.708475</td>\n",
       "      <td>8.108496</td>\n",
       "      <td>0.860854</td>\n",
       "      <td>0.498493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>55.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>65.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>75.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>95.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                age  time_in_hospital  num_procedures  num_medications  \\\n",
       "count  98052.000000      98052.000000    98052.000000     98052.000000   \n",
       "mean      66.260250          4.422011        1.350712        16.119590   \n",
       "std       15.590805          2.993070        1.708475         8.108496   \n",
       "min        5.000000          1.000000        0.000000         1.000000   \n",
       "25%       55.000000          2.000000        0.000000        11.000000   \n",
       "50%       65.000000          4.000000        1.000000        15.000000   \n",
       "75%       75.000000          6.000000        2.000000        20.000000   \n",
       "max       95.000000         14.000000        6.000000        81.000000   \n",
       "\n",
       "               race        gender  \n",
       "count  98052.000000  98052.000000  \n",
       "mean       1.658987      0.461174  \n",
       "std        0.860854      0.498493  \n",
       "min        0.000000      0.000000  \n",
       "25%        2.000000      0.000000  \n",
       "50%        2.000000      0.000000  \n",
       "75%        2.000000      1.000000  \n",
       "max        4.000000      1.000000  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "ddc0f3bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Many variables have a skewed distribution\n",
    "X.hist(figsize=(20,20));"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a21a2de",
   "metadata": {},
   "source": [
    "### Splitting Data into Train/Test Sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "57f3a5b2",
   "metadata": {
    "id": "57f3a5b2"
   },
   "outputs": [],
   "source": [
    "# Split the preprocessed data into a training and testing dataset\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.2, random_state= 42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9ddc5e1b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    69498\n",
       "1     8943\n",
       "Name: readmitted, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cd7fa827",
   "metadata": {
    "id": "cd7fa827"
   },
   "outputs": [],
   "source": [
    "# Create a StandardScaler instances\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Fit the StandardScaler\n",
    "X_scaler = scaler.fit(X_train)\n",
    "\n",
    "# Scale the data\n",
    "X_train_scaled = X_scaler.transform(X_train)\n",
    "X_test_scaled = X_scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2a2df944",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((78441, 6), (19611, 6), (78441,), (19611,))"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_scaled.shape, X_test_scaled.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35ab301e",
   "metadata": {},
   "source": [
    "We are using 19,611 data points, or 20% of the total, in our test set. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88ef8103",
   "metadata": {},
   "source": [
    "### Training and Comparing Different Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8d4cfba5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "4a6ed97f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize Classifiers\n",
    "tree= DecisionTreeClassifier(random_state=24) # using the random state for reproducibility\n",
    "forest= RandomForestClassifier(random_state=24)\n",
    "knn= KNeighborsClassifier()\n",
    "lg = LogisticRegression(fit_intercept=True, solver= 'liblinear', penalty='l1', random_state=24)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "cd83bb31",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a list with the objects \n",
    "models= [tree, forest, knn, lg]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "6e104202",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: DecisionTreeClassifier\n",
      "Time to Run: -0.0001230000000003173 seconds\n",
      "Train Score: 0.9149424408153899\n",
      "Test Score: 0.8595176176635562\n",
      "\n",
      "Model: RandomForestClassifier\n",
      "Time to Run: 0.0009117000000014031 seconds\n",
      "Train Score: 0.9149296923802603\n",
      "Test Score: 0.873897302534292\n",
      "\n",
      "Model: KNeighborsClassifier\n",
      "Time to Run: -0.00015739999999908605 seconds\n",
      "Train Score: 0.8869978710113333\n",
      "Test Score: 0.8828208658405997\n",
      "\n",
      "Model: LogisticRegression\n",
      "Time to Run: -0.001352300000000639 seconds\n",
      "Train Score: 0.885990744636096\n",
      "Test Score: 0.8917444291469073\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Loop through models to determine which model performs best on test set, with time measurements\n",
    "# Caution: SVM model takes X to run\n",
    "import timeit\n",
    "\n",
    "for model in models:\n",
    "    start = timeit.timeit()\n",
    "    model.fit(X_train_scaled, y_train) # fit the model\n",
    "    y_pred= model.predict(X_test) # then predict on the test set\n",
    "    end = timeit.timeit()\n",
    "    print(f'Model: {type(model).__name__}')\n",
    "    print(f'Time to Run: {end - start} seconds')\n",
    "    print(f'Train Score: {model.score(X_train_scaled, y_train)}')\n",
    "    print(f'Test Score: {model.score(X_test_scaled, y_test)}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c002402e",
   "metadata": {},
   "source": [
    "**The results show that Logistic Regression perform best on the test set, with a score of 89.** \n",
    "\n",
    "Now, let's view differences in accuracy, precision, recall and F1 scores between our models:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "df7cb621",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: DecisionTreeClassifier\n",
      "Accuracy is 0.52\n",
      "Precision is 0.11\n",
      "Recall is 0.50\n",
      "F1 is 0.19\n",
      "\n",
      "\n",
      "Model: RandomForestClassifier\n",
      "Accuracy is 0.88\n",
      "Precision is 0.13\n",
      "Recall is 0.01\n",
      "F1 is 0.02\n",
      "\n",
      "\n",
      "Model: KNeighborsClassifier\n",
      "Accuracy is 0.86\n",
      "Precision is 0.14\n",
      "Recall is 0.06\n",
      "F1 is 0.08\n",
      "\n",
      "\n",
      "Model: LogisticRegression\n",
      "Accuracy is 0.16\n",
      "Precision is 0.11\n",
      "Recall is 0.96\n",
      "F1 is 0.20\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Loop through models to evaluate precision, recall and F1 scores\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "for model in models:\n",
    "    model.fit(X_train_scaled, y_train) # fit the model\n",
    "    y_pred= model.predict(X_test) # then predict on the test set\n",
    "    print(f'Model: {type(model).__name__}')\n",
    "    print(\"Accuracy is {0:.2f}\".format(accuracy_score(y_test, y_pred)))\n",
    "    print(\"Precision is {0:.2f}\".format(precision_score(y_test, y_pred)))\n",
    "    print(\"Recall is {0:.2f}\".format(recall_score(y_test, y_pred)))\n",
    "    print(\"F1 is {0:.2f}\".format(f1_score(y_test, y_pred)))\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5959ba67",
   "metadata": {},
   "source": [
    "Although Logistic Regression has the lowest Accuracy score, Logistic Regression & Decision Tree Classifier models have the highest F1 score. Logistic Regression also has the highest recall score. Recall score is the measure of the correctly identified positive cases from all the actual positive cases. F1 score is defined as the harmonic mean of the model's precision and recall. It is a useful, balanced metric which gives a better measure of the incorrectly classified cases than the Accuracy Metric. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62b8eafd",
   "metadata": {},
   "source": [
    "### Compare model accuracy distribution using ShuffleSplit\n",
    "ShuffleSplit method cross validates x & y values for the entire set. Plotting a series of accuracy scores will give us a better understanding of the logistic regression and Decision Tree classifier's performance, compared to a single score calculated after a single trial. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b81c93c2",
   "metadata": {},
   "source": [
    "**In order to know if we really have the best model, we need to look at the distribution of the accuracy measures. We might want the best average model or the least variable model, depending on the situation.** "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3991d9f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies\n",
    "from sklearn.model_selection import ShuffleSplit\n",
    "from sklearn.model_selection import cross_validate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "042a4fef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a shuffle split instance to randomly spilt the dataset into 100 training and test pairs\n",
    "#  logistic regression\n",
    "rs1 = ShuffleSplit (n_splits=100, test_size=.03)\n",
    "cv_results_lg = cross_validate(\n",
    "    lg, X_test_scaled, y_test, cv=rs1, scoring='accuracy'\n",
    ")\n",
    "\n",
    "# Convert accuracy scores from list into series\n",
    "accuracy_scores_lg = pd.Series(cv_results_lg['test_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ba249538",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8897623089983022"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View resulting scores \n",
    "accuracy_scores_lg.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "be30b0b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Score: 0.89 [5th percentile: 0.869 & 95th percentile: 0.91]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot accuracy distributions using a box plot\n",
    "# logistic regression\n",
    "accuracy_scores_lg.plot(\n",
    "    title='Distribution of logistic regression classifier accuracy',\n",
    "    kind='box'\n",
    ")\n",
    "\n",
    "print(\n",
    "    'Average Score: {:.3} [5th percentile: {:.3} & 95th percentile: {:.3}]'.format(\n",
    "    accuracy_scores_lg.mean(),\n",
    "    accuracy_scores_lg.quantile(.05),\n",
    "    accuracy_scores_lg.quantile(.95),\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d487cef",
   "metadata": {},
   "source": [
    "This box plot shows the best (91%) and worst-case (87%) scenarios of our classifier's performance using ShuffleSplit and Cross Validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "41f48d93",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a shuffle split instance to randomly spilt the dataset into 100 training and test pairs\n",
    "#  decision tree\n",
    "rs2 = ShuffleSplit (n_splits=100, test_size=.03)\n",
    "cv_results_tree = cross_validate(\n",
    "    tree, X_test_scaled, y_test, cv=rs2, scoring='accuracy'\n",
    ")\n",
    "\n",
    "# Convert accuracy scores from list into series\n",
    "accuracy_scores_tree = pd.Series(cv_results_tree['test_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "7f0eccbf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8335483870967744"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View resulting scores \n",
    "accuracy_scores_tree.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b047649f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Score: 0.834 [5th percentile: 0.81 & 95th percentile: 0.856]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot accuracy's distribution using a box plot\n",
    "accuracy_scores_tree.plot(\n",
    "    title='Distribution of decision tree classifier accuracy',\n",
    "    kind='box'\n",
    ")\n",
    "\n",
    "print(\n",
    "    'Average Score: {:.3} [5th percentile: {:.3} & 95th percentile: {:.3}]'.format(\n",
    "    accuracy_scores_tree.mean(),\n",
    "    accuracy_scores_tree.quantile(.05),\n",
    "    accuracy_scores_tree.quantile(.95),\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1658bb35",
   "metadata": {},
   "source": [
    "This box plot shows the best (85%) and worst-case (82%) scenarios of our classifier's performance using ShuffleSplit and Cross Validation. \n",
    "\n",
    "**These results allow us to objectively say that the Logistic Regression model is the best model in our case.** "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b1b33e2",
   "metadata": {},
   "source": [
    "### HyperTuning the Logistic Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "07deb07a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['C', 'class_weight', 'dual', 'fit_intercept', 'intercept_scaling', 'l1_ratio', 'max_iter', 'multi_class', 'n_jobs', 'penalty', 'random_state', 'solver', 'tol', 'verbose', 'warm_start'])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View available parameters for logistic regression estimator\n",
    "lg.get_params().keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "74f78b15",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n",
      "[CV 1/5] END .......................penalty=l1, solver=lbfgs; total time=   0.0s\n",
      "[CV 2/5] END .......................penalty=l1, solver=lbfgs; total time=   0.0s\n",
      "[CV 3/5] END .......................penalty=l1, solver=lbfgs; total time=   0.0s\n",
      "[CV 4/5] END .......................penalty=l1, solver=lbfgs; total time=   0.0s\n",
      "[CV 5/5] END .......................penalty=l1, solver=lbfgs; total time=   0.0s\n",
      "[CV 1/5] END ...................penalty=l1, solver=newton-cg; total time=   0.0s\n",
      "[CV 2/5] END ...................penalty=l1, solver=newton-cg; total time=   0.0s\n",
      "[CV 3/5] END ...................penalty=l1, solver=newton-cg; total time=   0.0s\n",
      "[CV 4/5] END ...................penalty=l1, solver=newton-cg; total time=   0.0s\n",
      "[CV 5/5] END ...................penalty=l1, solver=newton-cg; total time=   0.0s\n",
      "[CV 1/5] END ...................penalty=l1, solver=liblinear; total time=   0.0s\n",
      "[CV 2/5] END ...................penalty=l1, solver=liblinear; total time=   0.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver lbfgs supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver lbfgs supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver lbfgs supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver lbfgs supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver lbfgs supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver newton-cg supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver newton-cg supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver newton-cg supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver newton-cg supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 443, in _check_solver\n",
      "    raise ValueError(\"Solver %s supports only 'l2' or 'none' penalties, \"\n",
      "ValueError: Solver newton-cg supports only 'l2' or 'none' penalties, got l1 penalty.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV 3/5] END ...................penalty=l1, solver=liblinear; total time=   0.0s\n",
      "[CV 4/5] END ...................penalty=l1, solver=liblinear; total time=   0.0s\n",
      "[CV 5/5] END ...................penalty=l1, solver=liblinear; total time=   0.0s\n",
      "[CV 1/5] END .......................penalty=l2, solver=lbfgs; total time=   0.0s\n",
      "[CV 2/5] END .......................penalty=l2, solver=lbfgs; total time=   0.0s\n",
      "[CV 3/5] END .......................penalty=l2, solver=lbfgs; total time=   0.0s\n",
      "[CV 4/5] END .......................penalty=l2, solver=lbfgs; total time=   0.0s\n",
      "[CV 5/5] END .......................penalty=l2, solver=lbfgs; total time=   0.0s\n",
      "[CV 1/5] END ...................penalty=l2, solver=newton-cg; total time=   0.1s\n",
      "[CV 2/5] END ...................penalty=l2, solver=newton-cg; total time=   0.0s\n",
      "[CV 3/5] END ...................penalty=l2, solver=newton-cg; total time=   0.1s\n",
      "[CV 4/5] END ...................penalty=l2, solver=newton-cg; total time=   0.0s\n",
      "[CV 5/5] END ...................penalty=l2, solver=newton-cg; total time=   0.1s\n",
      "[CV 1/5] END ...................penalty=l2, solver=liblinear; total time=   0.0s\n",
      "[CV 2/5] END ...................penalty=l2, solver=liblinear; total time=   0.0s\n",
      "[CV 3/5] END ...................penalty=l2, solver=liblinear; total time=   0.0s\n",
      "[CV 4/5] END ...................penalty=l2, solver=liblinear; total time=   0.0s\n",
      "[CV 5/5] END ...................penalty=l2, solver=liblinear; total time=   0.0s\n",
      "[CV 1/5] END ..............penalty=elastic_net, solver=lbfgs; total time=   0.0s\n",
      "[CV 2/5] END ..............penalty=elastic_net, solver=lbfgs; total time=   0.0s\n",
      "[CV 3/5] END ..............penalty=elastic_net, solver=lbfgs; total time=   0.0s\n",
      "[CV 4/5] END ..............penalty=elastic_net, solver=lbfgs; total time=   0.0s\n",
      "[CV 5/5] END ..............penalty=elastic_net, solver=lbfgs; total time=   0.0s\n",
      "[CV 1/5] END ..........penalty=elastic_net, solver=newton-cg; total time=   0.0s\n",
      "[CV 2/5] END ..........penalty=elastic_net, solver=newton-cg; total time=   0.0s\n",
      "[CV 3/5] END ..........penalty=elastic_net, solver=newton-cg; total time=   0.0s\n",
      "[CV 4/5] END ..........penalty=elastic_net, solver=newton-cg; total time=   0.0s\n",
      "[CV 5/5] END ..........penalty=elastic_net, solver=newton-cg; total time=   0.0s\n",
      "[CV 1/5] END ..........penalty=elastic_net, solver=liblinear; total time=   0.0s\n",
      "[CV 2/5] END ..........penalty=elastic_net, solver=liblinear; total time=   0.0s\n",
      "[CV 3/5] END ..........penalty=elastic_net, solver=liblinear; total time=   0.0s\n",
      "[CV 4/5] END ..........penalty=elastic_net, solver=liblinear; total time=   0.0s\n",
      "[CV 5/5] END ..........penalty=elastic_net, solver=liblinear; total time=   0.0s\n",
      "[CV 1/5] END .....................penalty=none, solver=lbfgs; total time=   0.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 439, in _check_solver\n",
      "    raise ValueError(\"Logistic Regression supports only penalties in %s,\"\n",
      "ValueError: Logistic Regression supports only penalties in ['l1', 'l2', 'elasticnet', 'none'], got elastic_net.\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV 2/5] END .....................penalty=none, solver=lbfgs; total time=   0.0s\n",
      "[CV 3/5] END .....................penalty=none, solver=lbfgs; total time=   0.0s\n",
      "[CV 4/5] END .....................penalty=none, solver=lbfgs; total time=   0.0s\n",
      "[CV 5/5] END .....................penalty=none, solver=lbfgs; total time=   0.0s\n",
      "[CV 1/5] END .................penalty=none, solver=newton-cg; total time=   0.0s\n",
      "[CV 2/5] END .................penalty=none, solver=newton-cg; total time=   0.1s\n",
      "[CV 3/5] END .................penalty=none, solver=newton-cg; total time=   0.0s\n",
      "[CV 4/5] END .................penalty=none, solver=newton-cg; total time=   0.0s\n",
      "[CV 5/5] END .................penalty=none, solver=newton-cg; total time=   0.1s\n",
      "[CV 1/5] END .................penalty=none, solver=liblinear; total time=   0.0s\n",
      "[CV 2/5] END .................penalty=none, solver=liblinear; total time=   0.0s\n",
      "[CV 3/5] END .................penalty=none, solver=liblinear; total time=   0.0s\n",
      "[CV 4/5] END .................penalty=none, solver=liblinear; total time=   0.0s\n",
      "[CV 5/5] END .................penalty=none, solver=liblinear; total time=   0.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 454, in _check_solver\n",
      "    raise ValueError(\n",
      "ValueError: penalty='none' is not supported for the liblinear solver\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 454, in _check_solver\n",
      "    raise ValueError(\n",
      "ValueError: penalty='none' is not supported for the liblinear solver\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 454, in _check_solver\n",
      "    raise ValueError(\n",
      "ValueError: penalty='none' is not supported for the liblinear solver\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 454, in _check_solver\n",
      "    raise ValueError(\n",
      "ValueError: penalty='none' is not supported for the liblinear solver\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py:610: FitFailedWarning: Estimator fit failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\", line 593, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 1306, in fit\n",
      "    solver = _check_solver(self.solver, self.penalty, self.dual)\n",
      "  File \"C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py\", line 454, in _check_solver\n",
      "    raise ValueError(\n",
      "ValueError: penalty='none' is not supported for the liblinear solver\n",
      "\n",
      "  warnings.warn(\"Estimator fit failed. The score on this train-test\"\n",
      "C:\\Users\\laure\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:918: UserWarning: One or more of the test scores are non-finite: [       nan        nan 0.88599074 0.88599074 0.88599074 0.88599074\n",
      "        nan        nan        nan 0.88599074 0.88599074        nan]\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(estimator=LogisticRegression(random_state=24),\n",
       "             param_grid={'penalty': ['l1', 'l2', 'elastic_net', 'none'],\n",
       "                         'solver': ['lbfgs', 'newton-cg', 'liblinear']},\n",
       "             verbose=4)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Run GridSearchCV on best model, in this case, logistic regression model\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Define model & parameters\n",
    "lg = LogisticRegression(fit_intercept=True, random_state =24)\n",
    "param_grid = {'solver': ['lbfgs', 'newton-cg', 'liblinear'], \n",
    "              'penalty': ['l1', 'l2', 'elastic_net', 'none'],}\n",
    "\n",
    "grid_lg = GridSearchCV(lg, param_grid, verbose=4)\n",
    "\n",
    "# Fit the model by using the grid search classifier. This will take the SVC model and try each combination of parameters.\n",
    "grid_lg.fit(X_train_scaled, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "06b036e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'penalty': 'l1', 'solver': 'liblinear'}"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# List best params\n",
    "grid_lg.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "1fe2e64d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8859907448870153"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " # List the best score\n",
    "grid_lg.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "01d5e506",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 1, ..., 1, 1, 1], dtype=int64)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " # Make predictions with the hypertuned model\n",
    "predictions = grid_lg.predict(X_test)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "17156e29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8917444291469073"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Score the hypertuned model on the test dataset\n",
    "grid_lg.score(X_test_scaled, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecc7507b",
   "metadata": {},
   "source": [
    "## Evaluating the Logistic Regression Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "719ef4dc",
   "metadata": {},
   "source": [
    "##### Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "lzTYC8mj_Mxx",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "lzTYC8mj_Mxx",
    "outputId": "0d939ef2-9e14-486f-a640-c85b7f0a34bc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   precision    recall  f1-score   support\n",
      "\n",
      "            other       0.93      0.06      0.11     17488\n",
      "early readmission       0.11      0.96      0.20      2123\n",
      "\n",
      "         accuracy                           0.16     19611\n",
      "        macro avg       0.52      0.51      0.15     19611\n",
      "     weighted avg       0.84      0.16      0.12     19611\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix , classification_report\n",
    "lg.fit(X_train_scaled, y_train)\n",
    "y_true = y_test\n",
    "y_pred = lg.predict(X_test)\n",
    "\n",
    "target_names = ['other', 'early readmission']\n",
    "print(classification_report(y_true, y_pred, target_names=target_names))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7bba2b1",
   "metadata": {
    "id": "9wtmX29vAGfs"
   },
   "source": [
    "The classification report shows that precision is high at 93% for Class 0, meaning that our Logistic Regression model was careful to avoid labeling things Class 0 that aren't Class 0. However, accuracy is very low at .16, showing that the performance of the model is poor. \n",
    "\n",
    "One reason may be the imbalance of the data, since there are many objects of class 0 and few samples of class 1. As a result, the classifier is overfitting to class 0 to minimize error during training. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0843a9c5",
   "metadata": {},
   "source": [
    "##### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "f7a720a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get and reshape confusion matrix data\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "matrix = confusion_matrix(y_true, y_pred)\n",
    "matrix = matrix.astype('float') / matrix.sum(axis=1)[:, np.newaxis]\n",
    "\n",
    "# Build the plot\n",
    "plt.figure(figsize=(16,7))\n",
    "sns.set(font_scale=2)\n",
    "sns.heatmap(matrix, annot=True, annot_kws={'size':20},\n",
    "            cmap=plt.cm.Greens, linewidths=0.2)\n",
    "\n",
    "# Add labels to the plot\n",
    "class_names = ['other', 'early readmission']\n",
    "tick_marks = np.arange(len(class_names))\n",
    "tick_marks2 = tick_marks + 0.5\n",
    "plt.xticks(tick_marks, class_names, rotation=0)\n",
    "plt.yticks(tick_marks2, class_names, rotation=0)\n",
    "plt.xlabel('Predicted label')\n",
    "plt.ylabel('True label')\n",
    "plt.title('Confusion Matrix for Logistic Regression Model')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3a56fb8",
   "metadata": {},
   "source": [
    "The confusion matrix shows that are many false positives (top right corner) and true negatives (bottom right corner) in our dataset. This means that our model predicted positive, but it's false. It also means that our model predicted negative, and it's true. This means that our model is not performing well. We want many True Positives and True Negatives. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b0dfc22",
   "metadata": {},
   "source": [
    "##### ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "356c3c67",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "0656ffe7",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_test\n",
    "y_pred_proba = lg.predict_proba(X_test)[:,1]\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_proba)\n",
    "roc_auc = auc(fpr, tpr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "0fb46a5f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(fpr, tpr, color='darkorange',\n",
    "         label=f'ROC curve (area = {roc_auc:0.2f})')\n",
    "plt.plot([0, 1], [0, 1], color='navy', linestyle='--')\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('ROC Curve - Logistic Regression')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6703cd8a",
   "metadata": {},
   "source": [
    "The further the curve is from the diagonal line, the better the model is at discriminating between positives and negatives in general. Because the line on our ROC curve is close to the diagonal line, the ROC curve that our logistic regression model does not discriminate between positives and negatives very well. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce7853c1",
   "metadata": {},
   "source": [
    "## Further Tweaks to our Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afeaa769",
   "metadata": {},
   "source": [
    "Since our target variable has a class imbalance problem, some methods we could use are: \n",
    "1. apply weights to predict a majority v. minority label\n",
    "2. up-sample or down-sample the training data to help balance the prediction\n",
    "3. use SMOTE\n",
    "5. We could also look at using a model that is better at handling imbalanced data, such as machine learning/neural networks, or a classical algorithm such as bagging or boosting algorithms and/or decision trees"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "516586a5",
   "metadata": {},
   "source": [
    "## Applying SMOTE"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f1eddc8",
   "metadata": {},
   "source": [
    "SMOTE first selects a minority class instance a at random and finds its k nearest minority class neighbors. The synthetic instance is then created by choosing one of the k nearest neighbors b at random and connecting a and b to form a line segment in the feature space. The synthetic instances are generated as a convex combination of the two chosen instances a and b.\n",
    "\n",
    "Source: https://machinelearningmastery.com/smote-oversampling-for-imbalanced-classification/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "fba697a2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already up-to-date: imbalanced-learn in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (0.8.1)\n",
      "Requirement already satisfied, skipping upgrade: scipy>=0.19.1 in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (from imbalanced-learn) (1.5.0)\n",
      "Requirement already satisfied, skipping upgrade: numpy>=1.13.3 in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (from imbalanced-learn) (1.18.5)\n",
      "Requirement already satisfied, skipping upgrade: joblib>=0.11 in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (from imbalanced-learn) (1.0.1)\n",
      "Requirement already satisfied, skipping upgrade: scikit-learn>=0.24 in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (from imbalanced-learn) (0.24.2)\n",
      "Requirement already satisfied, skipping upgrade: threadpoolctl>=2.0.0 in c:\\users\\laure\\anaconda3\\envs\\pythondata\\lib\\site-packages (from scikit-learn>=0.24->imbalanced-learn) (2.2.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install -U imbalanced-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "d1f86ee5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.over_sampling import SMOTE\n",
    "from collections import Counter # Counter object summarizes the number of examples in each class "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "6f975384",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original dataset shape Counter({0: 69498, 1: 8943})\n",
      "New dataset shape Counter({0: 69498, 1: 69498})\n"
     ]
    }
   ],
   "source": [
    "print('Original dataset shape {}'.format(Counter(y_train)))\n",
    "X_resampled, y_resampled = SMOTE(random_state=24).fit_resample(X_train, y_train)\n",
    "print('New dataset shape {}'.format(Counter(y_resampled)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75853caa",
   "metadata": {},
   "source": [
    "### SMOTE - Logistic Regression v.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "94243f3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "X_resampled = pd.DataFrame(X_resampled, columns = list(X.columns))\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_resampled, y_resampled, \n",
    "                                                    test_size=0.20, random_state=42)\n",
    "lg2 = LogisticRegression(fit_intercept=True, penalty='l2')\n",
    "lg2.fit(X_train, y_train)\n",
    "\n",
    "logit_pred = lg2.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "811264f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy is 0.56\n",
      "Precision is 0.55\n",
      "Recall is 0.58\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "print(\"Accuracy is {0:.2f}\".format(accuracy_score(y_test, logit_pred)))\n",
    "print(\"Precision is {0:.2f}\".format(precision_score(y_test, logit_pred)))\n",
    "print(\"Recall is {0:.2f}\".format(recall_score(y_test, logit_pred)))\n",
    "\n",
    "accuracy_logit = accuracy_score(y_test, logit_pred)\n",
    "precision_logit = precision_score(y_test, logit_pred)\n",
    "recall_logit = recall_score(y_test, logit_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74747ab1",
   "metadata": {},
   "source": [
    "### SMOTE - Decision Tree v.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "ae8a7f20",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "dtree = DecisionTreeClassifier(max_depth=28, criterion = \"entropy\", min_samples_split=10)\n",
    "dtree.fit(X_train, y_train)\n",
    "\n",
    "dtree_pred = dtree.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "ab649727",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy is 0.72\n",
      "Precision is 0.70\n",
      "Recall is 0.76\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy is {0:.2f}\".format(accuracy_score(y_test, dtree_pred)))\n",
    "print(\"Precision is {0:.2f}\".format(precision_score(y_test, dtree_pred)))\n",
    "print(\"Recall is {0:.2f}\".format(recall_score(y_test, dtree_pred)))\n",
    "\n",
    "accuracy_dtree = accuracy_score(y_test, dtree_pred)\n",
    "precision_dtree = precision_score(y_test, dtree_pred)\n",
    "recall_dtree = recall_score(y_test, dtree_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f767f964",
   "metadata": {},
   "source": [
    "### SMOTE - Random Forest v.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "7b11fedd",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rm = RandomForestClassifier()\n",
    "rm.fit(X_train, y_train)\n",
    "rm_prd = rm.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "416d7dde",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy is 0.74\n",
      "Precision is 0.71\n",
      "Recall is 0.80\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy is {0:.2f}\".format(accuracy_score(y_test, rm_prd)))\n",
    "print(\"Precision is {0:.2f}\".format(precision_score(y_test, rm_prd)))\n",
    "print(\"Recall is {0:.2f}\".format(recall_score(y_test, rm_prd)))\n",
    "\n",
    "accuracy_rm = accuracy_score(y_test, rm_prd)\n",
    "precision_rm = precision_score(y_test, rm_prd)\n",
    "recall_rm = recall_score(y_test, rm_prd)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fced7f4",
   "metadata": {},
   "source": [
    "### SMOTE - KNN v.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "624632bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn2 = KNeighborsClassifier()\n",
    "knn2.fit(X_train, y_train)\n",
    "knn2_prd = knn2.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "609efeb0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy is 0.70\n",
      "Precision is 0.67\n",
      "Recall is 0.77\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy is {0:.2f}\".format(accuracy_score(y_test, knn2_prd)))\n",
    "print(\"Precision is {0:.2f}\".format(precision_score(y_test, knn2_prd)))\n",
    "print(\"Recall is {0:.2f}\".format(recall_score(y_test, knn2_prd)))\n",
    "\n",
    "accuracy_knn2 = accuracy_score(y_test, knn2_prd)\n",
    "precision_knn2 = precision_score(y_test, knn2_prd)\n",
    "recall_knn2 = recall_score(y_test, knn2_prd)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a43b279",
   "metadata": {},
   "source": [
    "# Comparing Models with SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "3011ca77",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-63-3cb4545a5393>:12: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  ax = plt.subplot(111)\n",
      "<ipython-input-63-3cb4545a5393>:21: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  ax = plt.subplot(111)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_palette(\"hls\")\n",
    "plt.figure(figsize=(14, 7))\n",
    "ax = plt.subplot(111)\n",
    "\n",
    "models = ['Logistic Regression', 'Decision Tree', 'Random Forest', 'KNN']\n",
    "values = [accuracy_logit, accuracy_dtree, accuracy_rm, accuracy_knn2]\n",
    "model = np.arange(len(models))\n",
    "\n",
    "plt.bar(model, values, align='center', width = 0.15, alpha=0.7, color = 'orange', label= 'accuracy')\n",
    "plt.xticks(model, models)\n",
    "           \n",
    "ax = plt.subplot(111)\n",
    "\n",
    "models = ['Logistic Regression', 'Decision Tree', 'Random Forest', 'KNN']\n",
    "values = [precision_logit, precision_dtree, precision_rm, precision_knn2]\n",
    "model = np.arange(len(models))\n",
    "\n",
    "plt.bar(model+0.15, values, align='center', width = 0.15, alpha=0.7, color = 'lightgreen', label = 'precision')\n",
    "plt.xticks(model, models)\n",
    "\n",
    "ax = plt.subplot(111)\n",
    "\n",
    "models = ['Logistic Regression', 'Decision Tree', 'Random Forest', 'KNN']\n",
    "values = [recall_logit, recall_dtree, recall_rm, recall_knn2]\n",
    "model = np.arange(len(models))\n",
    "\n",
    "plt.bar(model+0.3, values, align='center', width = 0.15, alpha=0.7, color = 'lightblue', label = 'recall')\n",
    "plt.xticks(model, models)\n",
    "\n",
    "plt.ylabel('Performance Metrics')\n",
    "plt.xlabel('Models')\n",
    "plt.title('Model Comparison - Using SMOTE')\n",
    "    \n",
    "# removing the axis on the top and right of the plot window\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.legend()\n",
    "\n",
    "plt.show();           "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c5654ff",
   "metadata": {},
   "source": [
    "## Evaluating our Winning Model - Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7df82138",
   "metadata": {},
   "source": [
    "### Feature Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "2a44adda",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Evaluate feature performance using random forest model\n",
    "important_features = pd.DataFrame({'Features': X_train.columns, \n",
    "                                   'Importance': rm.feature_importances_})\n",
    "\n",
    "# sort the dataframe in the descending order according to the feature importance\n",
    "important_features = important_features.sort_values('Importance', ascending = False)\n",
    "\n",
    "# create a barplot to visualize the features based on their importance\n",
    "sns.set(font_scale=2)\n",
    "sns.set_palette(\"hls\")\n",
    "plt.figure(figsize=(12,12))\n",
    "sns.barplot(x = 'Importance', y = 'Features', data = important_features)\n",
    "\n",
    "# add plot and axes labels\n",
    "# set text size using 'fontsize'\n",
    "plt.title('Feature Importance')\n",
    "plt.xlabel('Importance')\n",
    "plt.ylabel('Features')\n",
    "\n",
    "# display the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ceb53e3c",
   "metadata": {},
   "source": [
    "Number of medications, time in hospital and age features carry the most importance.  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e63a17dd",
   "metadata": {},
   "source": [
    "### Confusion Matrix "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "4ee019ed",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get and reshape confusion matrix data\n",
    "from sklearn.metrics import confusion_matrix\n",
    "matrix = confusion_matrix(y_test, rm_prd)\n",
    "matrix = matrix.astype('float') / matrix.sum(axis=1)[:, np.newaxis]\n",
    "\n",
    "# Build the plot\n",
    "plt.figure(figsize=(16,7))\n",
    "sns.set(font_scale=2)\n",
    "sns.heatmap(matrix, annot=True, annot_kws={'size':20},\n",
    "            cmap=plt.cm.Blues, linewidths=0.2)\n",
    "\n",
    "# Add labels to the plot\n",
    "class_names = ['other', 'early readmission']\n",
    "tick_marks = np.arange(len(class_names))\n",
    "tick_marks2 = tick_marks + 0.5\n",
    "plt.xticks(tick_marks, class_names, rotation=0)\n",
    "plt.yticks(tick_marks2, class_names, rotation=0)\n",
    "plt.xlabel('Predicted label')\n",
    "plt.ylabel('True label')\n",
    "plt.title('Confusion Matrix for Random Forest Model')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "444054b9",
   "metadata": {},
   "source": [
    "### ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "f6a16e88",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "ae6a459e",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_test\n",
    "y_pred_proba = rm.predict_proba(X_test)[:,1]\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_proba)\n",
    "roc_auc = auc(fpr, tpr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "150bcc7f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(fpr, tpr, color='darkorange',\n",
    "         label=f'ROC curve (area = {roc_auc:0.2f})')\n",
    "plt.plot([0, 1], [0, 1], color='navy', linestyle='--')\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('ROC Curve Random Forest')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "456c5304",
   "metadata": {},
   "source": [
    "### Hypertuning the Random Forest Model\n",
    "Defining a grid of hyperparameter ranges, and randomly sampling from the grid, performing K-Fold CV for each combination of values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "3941101e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['bootstrap', 'ccp_alpha', 'class_weight', 'criterion', 'max_depth', 'max_features', 'max_leaf_nodes', 'max_samples', 'min_impurity_decrease', 'min_impurity_split', 'min_samples_leaf', 'min_samples_split', 'min_weight_fraction_leaf', 'n_estimators', 'n_jobs', 'oob_score', 'random_state', 'verbose', 'warm_start'])"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View available parameters for logistic regression estimator\n",
    "rm.get_params().keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bd3708e",
   "metadata": {},
   "source": [
    "Find Best Estimators for New RF Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "bf3fac63",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=3, estimator=RandomForestClassifier(),\n",
       "             param_grid={'criterion': ['gini', 'entropy'], 'max_depth': [50],\n",
       "                         'max_features': ['auto', 'sqrt', 0.3],\n",
       "                         'n_estimators': [500]},\n",
       "             verbose=3)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "# Create the parameter grid based on the results of random search \n",
    "param_grid = {\n",
    "    'max_depth': [25],\n",
    "    'criterion': ['gini', 'entropy'],\n",
    "    'max_features': ['auto', 'sqrt', .3],\n",
    "    'n_estimators': [500]\n",
    "}\n",
    "\n",
    "# Instantiate the grid search model\n",
    "grid_rm = GridSearchCV(estimator = rm, cv = 3, param_grid = param_grid, verbose = 3)\n",
    "grid_rm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "7b6f2e76",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 6 candidates, totalling 18 fits\n",
      "[CV 1/3] END criterion=gini, max_depth=50, max_features=auto, n_estimators=500; total time=  27.5s\n",
      "[CV 2/3] END criterion=gini, max_depth=50, max_features=auto, n_estimators=500; total time=  27.0s\n",
      "[CV 3/3] END criterion=gini, max_depth=50, max_features=auto, n_estimators=500; total time=  25.8s\n",
      "[CV 1/3] END criterion=gini, max_depth=50, max_features=sqrt, n_estimators=500; total time=  25.7s\n",
      "[CV 2/3] END criterion=gini, max_depth=50, max_features=sqrt, n_estimators=500; total time=  25.3s\n",
      "[CV 3/3] END criterion=gini, max_depth=50, max_features=sqrt, n_estimators=500; total time=  26.0s\n",
      "[CV 1/3] END criterion=gini, max_depth=50, max_features=0.3, n_estimators=500; total time=  21.2s\n",
      "[CV 2/3] END criterion=gini, max_depth=50, max_features=0.3, n_estimators=500; total time=  20.9s\n",
      "[CV 3/3] END criterion=gini, max_depth=50, max_features=0.3, n_estimators=500; total time=  20.7s\n",
      "[CV 1/3] END criterion=entropy, max_depth=50, max_features=auto, n_estimators=500; total time=  26.1s\n",
      "[CV 2/3] END criterion=entropy, max_depth=50, max_features=auto, n_estimators=500; total time=  25.9s\n",
      "[CV 3/3] END criterion=entropy, max_depth=50, max_features=auto, n_estimators=500; total time=  27.7s\n",
      "[CV 1/3] END criterion=entropy, max_depth=50, max_features=sqrt, n_estimators=500; total time=  27.7s\n",
      "[CV 2/3] END criterion=entropy, max_depth=50, max_features=sqrt, n_estimators=500; total time=  27.2s\n",
      "[CV 3/3] END criterion=entropy, max_depth=50, max_features=sqrt, n_estimators=500; total time=  27.0s\n",
      "[CV 1/3] END criterion=entropy, max_depth=50, max_features=0.3, n_estimators=500; total time=  24.0s\n",
      "[CV 2/3] END criterion=entropy, max_depth=50, max_features=0.3, n_estimators=500; total time=  22.7s\n",
      "[CV 3/3] END criterion=entropy, max_depth=50, max_features=0.3, n_estimators=500; total time=  26.0s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=3, estimator=RandomForestClassifier(),\n",
       "             param_grid={'criterion': ['gini', 'entropy'], 'max_depth': [50],\n",
       "                         'max_features': ['auto', 'sqrt', 0.3],\n",
       "                         'n_estimators': [500]},\n",
       "             verbose=3)"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fit the grid search to the data to try each combination of parameters\n",
    "grid_rm.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "d59ea4f1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'criterion': 'gini', 'max_depth': 50, 'max_features': 'auto', 'n_estimators': 500}\n"
     ]
    }
   ],
   "source": [
    "# List the best parameters for this dataset\n",
    "print(grid_rm.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "fda32cb4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7253948118358574"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# List the best score\n",
    "grid_rm.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "a29927a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 1, ..., 0, 1, 0], dtype=int64)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Make predictions with the hypertuned model\n",
    "predictions = grid_rm.predict(X_test)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "a698005f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7407553956834533"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Score the hypertuned model on the test dataset\n",
    "grid_accuracy = grid_rm.score(X_test, y_test)\n",
    "grid_accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "1c0663df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Improvement of 0.07%.\n"
     ]
    }
   ],
   "source": [
    "# View Improvement\n",
    "print('Improvement of {:0.2f}%.'.format( 100 * (grid_accuracy - accuracy_rm) / accuracy_rm))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60f2b255",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "599fabe9",
   "metadata": {},
   "source": [
    "### Plot accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df12d8c3",
   "metadata": {},
   "source": [
    "ShuffleSplit method cross validates x & y values for the entire set. Plotting a series of accuracy scores will give us a better understanding of the logistic regression and Decision Tree classifier's performance, compared to a single score calculated after a single trial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "78fdf962",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies\n",
    "from sklearn.model_selection import ShuffleSplit\n",
    "from sklearn.model_selection import cross_validate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "9e935225",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a shuffle split instance to randomly spilt the dataset into 100 training and test pairs\n",
    "#  logistic regression\n",
    "rm3 = ShuffleSplit (n_splits=50, test_size=.02)\n",
    "\n",
    "cv_results_rm = cross_validate(\n",
    "    rm, X_test, y_test, cv=rm3, scoring='accuracy'\n",
    ")\n",
    "\n",
    "# Convert accuracy scores from list into series\n",
    "accuracy_scores_rm3 = pd.Series(cv_results_rm['test_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "4db720c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Score: 0.676 [5th percentile: 0.644 & 95th percentile: 0.703]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot accuracy distributions using a box plot\n",
    "# logistic regression\n",
    "accuracy_scores_rm3.plot(\n",
    "    title='Distribution of Accuracy Across 50 Splits',\n",
    "    kind='box'\n",
    ")\n",
    "\n",
    "print(\n",
    "    'Average Score: {:.3} [5th percentile: {:.3} & 95th percentile: {:.3}]'.format(\n",
    "    accuracy_scores_rm3.mean(),\n",
    "    accuracy_scores_rm3.quantile(.05),\n",
    "    accuracy_scores_rm3.quantile(.95),\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acc279cd",
   "metadata": {},
   "source": [
    "## Export Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "WsPVDZhuAf-v",
   "metadata": {
    "id": "WsPVDZhuAf-v"
   },
   "outputs": [],
   "source": [
    "# Export our model to HDF5 file\n",
    "import joblib"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5669f121",
   "metadata": {},
   "source": [
    "Export Random Forest model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "96bb22ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/rm_model.h5']"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(rm, 'models/rm_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "9873cce1",
   "metadata": {},
   "outputs": [],
   "source": [
    "rm = joblib.load('models/rm_model.h5')"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "include_colab_link": true,
   "name": "diabetesv3.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}