risk_score_SVC.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(858, 36)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#################################\n",
    "###### 1. Data Preparation ######\n",
    "#################################\n",
    "\n",
    "# https://www.kaggle.com/loveall/cervical-cancer-risk-classification\n",
    "# https://ieeexplore.ieee.org/document/\n",
    "# NOTE: CHANGE ? to NA so that it can then compute mean and std for numeric cols\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.axes_grid1 import AxesGrid\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "df = pd.read_csv(\"kag_risk_factors_cervical_cancer.csv\")\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0., 1.]), array([803,  55]))"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(df['Biopsy'],return_counts=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data examination:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age                          26.820513\n",
       "STDs: Number of diagnosis     0.087413\n",
       "Dx:Cancer                     0.020979\n",
       "Dx:CIN                        0.010490\n",
       "Dx:HPV                        0.020979\n",
       "Dx                            0.027972\n",
       "Hinselmann                    0.040793\n",
       "Schiller                      0.086247\n",
       "Citology                      0.051282\n",
       "Biopsy                        0.064103\n",
       "dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "######### column names #########\n",
    "# print(df.columns)\n",
    "    # note: CIN - Cervical Intraepithelial Neoplasia\n",
    "    # note: Dx - Medical Abbreviation for Diagnosis\n",
    "\n",
    "######### other data descriptors #########\n",
    "# print(df.head(5))\n",
    "# df.describe\n",
    "# df.info()\n",
    "\n",
    "######### how many unique values per feature? #########\n",
    "# for col in df.columns:\n",
    "#     print(col,\": \")\n",
    "#     print(np.unique(df[col],return_counts=True))\n",
    "######### object type in each col: #########\n",
    "# for col in df.columns:\n",
    "    # print(df[col].dtype) # objects = string (meaning that because there's at least one string in that col, whole col treated\n",
    "        # as string, even if most of the other rows in that col are actually int!)\n",
    "    \n",
    "######### how many unknowns per feature? #########\n",
    "# for col in df.columns:\n",
    "#     print(f'{col} : {len(df[df[col]==\"?\"])}')\n",
    "\n",
    "######### mean and sd for each feature (ie column) #########\n",
    "df.mean()\n",
    "# df.std()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From Kaggle notebook (https://www.kaggle.com/atakansoztekin/cancer-data-analysis-and-model-implementation):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age                                     0\n",
       "Number of sexual partners              26\n",
       "First sexual intercourse                7\n",
       "Num of pregnancies                     56\n",
       "Smokes                                 13\n",
       "Smokes (years)                         13\n",
       "Smokes (packs/year)                    13\n",
       "Hormonal Contraceptives               108\n",
       "Hormonal Contraceptives (years)       108\n",
       "IUD                                   117\n",
       "IUD (years)                           117\n",
       "STDs                                  105\n",
       "STDs (number)                         105\n",
       "STDs:condylomatosis                   105\n",
       "STDs:cervical condylomatosis          105\n",
       "STDs:vaginal condylomatosis           105\n",
       "STDs:vulvo-perineal condylomatosis    105\n",
       "STDs:syphilis                         105\n",
       "STDs:pelvic inflammatory disease      105\n",
       "STDs:genital herpes                   105\n",
       "STDs:molluscum contagiosum            105\n",
       "STDs:AIDS                             105\n",
       "STDs:HIV                              105\n",
       "STDs:Hepatitis B                      105\n",
       "STDs:HPV                              105\n",
       "STDs: Number of diagnosis               0\n",
       "STDs: Time since first diagnosis      787\n",
       "STDs: Time since last diagnosis       787\n",
       "Dx:Cancer                               0\n",
       "Dx:CIN                                  0\n",
       "Dx:HPV                                  0\n",
       "Dx                                      0\n",
       "Hinselmann                              0\n",
       "Schiller                                0\n",
       "Citology                                0\n",
       "Biopsy                                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# replace \"?\" with na (easier to handle subsequently)\n",
    "df = df.replace(\"?\", np.nan)\n",
    "# convert to float\n",
    "df = df.astype(float)\n",
    "# how many null values per feature?\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(858, 34)\n"
     ]
    }
   ],
   "source": [
    "# remove features with too many null values\n",
    "df = df.drop(['STDs: Time since first diagnosis', 'STDs: Time since last diagnosis'], axis=1)\n",
    "# print(df.columns)\n",
    "print(df.shape) # should have 2 features less"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "######### Inferring missing data #########\n",
    "\n",
    "# Strategy 1: KNN\n",
    "from sklearn.impute import KNNImputer\n",
    "imputer = KNNImputer(n_neighbors=600)\n",
    "df = pd.DataFrame(imputer.fit_transform(df),columns=df.columns)\n",
    "# df.isnull().sum()\n",
    "\n",
    "# Strategy 2: replace na values with median of that feature\n",
    "    # (not mean, since mean more vulnerable to outliers)\n",
    "# for col in df.columns:\n",
    "#     df[col].fillna(df[col].median(), inplace = True)\n",
    "# df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['Age', 'Number of sexual partners', 'First sexual intercourse',\n",
      "       'Num of pregnancies', 'Smokes', 'Smokes (years)', 'Smokes (packs/year)',\n",
      "       'Hormonal Contraceptives', 'Hormonal Contraceptives (years)', 'IUD',\n",
      "       'IUD (years)', 'STDs', 'STDs (number)', 'STDs:condylomatosis',\n",
      "       'STDs:cervical condylomatosis', 'STDs:vaginal condylomatosis',\n",
      "       'STDs:vulvo-perineal condylomatosis', 'STDs:syphilis',\n",
      "       'STDs:pelvic inflammatory disease', 'STDs:genital herpes',\n",
      "       'STDs:molluscum contagiosum', 'STDs:AIDS', 'STDs:HIV',\n",
      "       'STDs:Hepatitis B', 'STDs:HPV', 'STDs: Number of diagnosis',\n",
      "       'Dx:Cancer', 'Dx:CIN', 'Dx:HPV', 'Dx', 'Hinselmann', 'Schiller',\n",
      "       'Citology', 'Biopsy'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Number of sexual partners</th>\n",
       "      <th>First sexual intercourse</th>\n",
       "      <th>Num of pregnancies</th>\n",
       "      <th>Smokes</th>\n",
       "      <th>Hormonal Contraceptives</th>\n",
       "      <th>Hormonal Contraceptives (years)</th>\n",
       "      <th>IUD</th>\n",
       "      <th>IUD (years)</th>\n",
       "      <th>STDs</th>\n",
       "      <th>...</th>\n",
       "      <th>STDs:HPV</th>\n",
       "      <th>STDs: Number of diagnosis</th>\n",
       "      <th>Dx:Cancer</th>\n",
       "      <th>Dx:CIN</th>\n",
       "      <th>Dx:HPV</th>\n",
       "      <th>Dx</th>\n",
       "      <th>Hinselmann</th>\n",
       "      <th>Schiller</th>\n",
       "      <th>Citology</th>\n",
       "      <th>Biopsy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>15.00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>15.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14.00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>34.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.55</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>52.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>16.00</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1369.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>46.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>21.00</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>23.00</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>51.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>17.00</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>115.600000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>26.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>26.00</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>45.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.00</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>44.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>15.00</td>\n",
       "      <td>2.473333</td>\n",
       "      <td>3.547524</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.131667</td>\n",
       "      <td>0.604117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age  Number of sexual partners  First sexual intercourse  \\\n",
       "0  18.0                        4.0                     15.00   \n",
       "1  15.0                        1.0                     14.00   \n",
       "2  34.0                        1.0                     17.55   \n",
       "3  52.0                        5.0                     16.00   \n",
       "4  46.0                        3.0                     21.00   \n",
       "5  42.0                        3.0                     23.00   \n",
       "6  51.0                        3.0                     17.00   \n",
       "7  26.0                        1.0                     26.00   \n",
       "8  45.0                        1.0                     20.00   \n",
       "9  44.0                        3.0                     15.00   \n",
       "\n",
       "   Num of pregnancies       Smokes  Hormonal Contraceptives  \\\n",
       "0            1.000000     0.000000                      0.0   \n",
       "1            1.000000     0.000000                      0.0   \n",
       "2            1.000000     0.000000                      0.0   \n",
       "3            4.000000  1369.000000                      1.0   \n",
       "4            4.000000     0.000000                      1.0   \n",
       "5            2.000000     0.000000                      0.0   \n",
       "6            6.000000   115.600000                      0.0   \n",
       "7            3.000000     0.000000                      1.0   \n",
       "8            5.000000     0.000000                      0.0   \n",
       "9            2.473333     3.547524                      0.0   \n",
       "\n",
       "   Hormonal Contraceptives (years)       IUD  IUD (years)  STDs  ...  \\\n",
       "0                              0.0  0.000000     0.000000   0.0  ...   \n",
       "1                              0.0  0.000000     0.000000   0.0  ...   \n",
       "2                              0.0  0.000000     0.000000   0.0  ...   \n",
       "3                              3.0  0.000000     0.000000   0.0  ...   \n",
       "4                             15.0  0.000000     0.000000   0.0  ...   \n",
       "5                              0.0  0.000000     0.000000   0.0  ...   \n",
       "6                              0.0  1.000000     7.000000   0.0  ...   \n",
       "7                              2.0  1.000000     7.000000   0.0  ...   \n",
       "8                              0.0  0.000000     0.000000   0.0  ...   \n",
       "9                              0.0  0.131667     0.604117   0.0  ...   \n",
       "\n",
       "   STDs:HPV  STDs: Number of diagnosis  Dx:Cancer  Dx:CIN  Dx:HPV   Dx  \\\n",
       "0       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "1       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "2       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "3       0.0                        0.0        1.0     0.0     1.0  0.0   \n",
       "4       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "5       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "6       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "7       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "8       0.0                        0.0        1.0     0.0     1.0  1.0   \n",
       "9       0.0                        0.0        0.0     0.0     0.0  0.0   \n",
       "\n",
       "   Hinselmann  Schiller  Citology  Biopsy  \n",
       "0         0.0       0.0       0.0     0.0  \n",
       "1         0.0       0.0       0.0     0.0  \n",
       "2         0.0       0.0       0.0     0.0  \n",
       "3         0.0       0.0       0.0     0.0  \n",
       "4         0.0       0.0       0.0     0.0  \n",
       "5         0.0       0.0       0.0     0.0  \n",
       "6         1.0       1.0       0.0     1.0  \n",
       "7         0.0       0.0       0.0     0.0  \n",
       "8         0.0       0.0       0.0     0.0  \n",
       "9         0.0       0.0       0.0     0.0  \n",
       "\n",
       "[10 rows x 32 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "######### Feature selection: smoking #########\n",
    "# how many cigarettes has this person smoked in their life?\n",
    "df['Smokes'] = df['Smokes (years)']*df['Smokes (packs/year)']\n",
    "# remove redundant columns\n",
    "df = df.drop(['Smokes (years)','Smokes (packs/year)'], axis=1)\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "######### Feature selection: contraceptives #########\n",
    "# remove redundant columns\n",
    "df = df.drop(['IUD','Hormonal Contraceptives','STDs'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "######### Normalize df #########\n",
    "# df=(df-df.mean())/df.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "######### normalized feature correlation matrix #########\n",
    "df_corr = df.corr()\n",
    "fig= plt.figure(figsize=(15,15))\n",
    "import seaborn as sns\n",
    "ax = sns.heatmap(df_corr, cmap='YlGnBu')\n",
    "plt.savefig(\"feature_corr_matrix.jpg\",bbox_inches = \"tight\")\n",
    "pd.DataFrame(df_corr).to_csv(\"feature_correlations.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "###### bootstrap example ######\n",
    "\n",
    "# from sklearn.utils import resample\n",
    "# df2 = resample(df[df['Biopsy']==1],n_samples=50,replace=True)\n",
    "# df = df.append(df2)\n",
    "# print(np.unique(df['Biopsy'],return_counts=True))\n",
    "# df.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "###### same proportion of each class in train and test sets ######\n",
    "\n",
    "df_biopsy_1 = df[df['Biopsy']==1]\n",
    "df_biopsy_0 = df[df['Biopsy']==0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_0 = df_biopsy_0.iloc[0:642,:]\n",
    "train_1 = df_biopsy_1.iloc[0:44,:]\n",
    "\n",
    "test_0 = df_biopsy_0.iloc[642:len(df_biopsy_0),:]\n",
    "test_1 = df_biopsy_1.iloc[44:len(df_biopsy_1),:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "642\n"
     ]
    }
   ],
   "source": [
    "###### boot strap class 1 to training data ######\n",
    "    # (difficult to use class weights with SVC for some reason)\n",
    "from sklearn.utils import resample\n",
    "n = len(train_0) - len(train_1)\n",
    "train_1_supp = resample(train_1,n_samples=n,replace=True)\n",
    "train_1 = train_1.append(train_1_supp)\n",
    "print(len(train_1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = train_0.append(train_1)\n",
    "test = test_0.append(test_1)\n",
    "# randomly reshuffle the rows\n",
    "train = train.sample(frac = 1).reset_index(drop=True)\n",
    "test = test.sample(frac = 1).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "###### preparing outputs for model ######\n",
    "\n",
    "# df1.iloc[,28] = Biopsy; df1.iloc[,27] = Citology; \n",
    "    # df1.iloc[,26] = Schiller; df1.iloc[,25] = Hinselmann\n",
    "    \n",
    "X_train = train.iloc[:,0:25] # 0 included, 687 excluded; 0 included, 25 excluded\n",
    "y_train_Hinselmann = train.iloc[:,25]\n",
    "y_train_Schiller = train.iloc[:,26]\n",
    "y_train_Citology = train.iloc[:,27]\n",
    "y_train_Biopsy = train.iloc[:,28]\n",
    "\n",
    "X_test = test.iloc[:,0:25] # 687 included, len(df1) excluded; 0 included, 25 excluded\n",
    "y_test_Hinselmann = test.iloc[:,25]\n",
    "y_test_Schiller = test.iloc[:,26]\n",
    "y_test_Citology = test.iloc[:,27]\n",
    "y_test_Biopsy = test.iloc[:,28]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimators: {'C': 0.05, 'loss': 'squared_hinge', 'penalty': 'l2'}\n",
      "(172,)\n",
      "<class 'numpy.ndarray'>\n",
      "balanced accuracy score:  0.6403162055335968\n"
     ]
    }
   ],
   "source": [
    "#########################################\n",
    "### 2.1 Linear SVC: Biopsy Prediction ###\n",
    "#########################################\n",
    "\n",
    "# packages\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.pipeline import Pipeline  \n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "SVC = GridSearchCV(estimator=LinearSVC() , param_grid={'penalty':['l1','l2'],'loss':['hinge','squared_hinge'],'C':[0.01,0.05,0.1,0.5,1]},cv=5, n_jobs=-1) # 5-fold CV\n",
    "\n",
    "SVC.fit(X_train, y_train_Biopsy)\n",
    "\n",
    "print(\"Best estimators: \" + str(SVC.best_params_))\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "y_pred = SVC.predict(X_test)\n",
    "print(y_pred.shape)\n",
    "print(type(y_pred))\n",
    "\n",
    "#>>> y_pred = [0, 1, 0, 0, 0, 1]\n",
    "print(\"balanced accuracy score: \",balanced_accuracy_score(y_test_Biopsy, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Remember outputs:\n",
    "Best estimators: {'C': 0.05, 'loss': 'squared_hinge', 'penalty': 'l2'}\n",
    "(172,)\n",
    "<class 'numpy.ndarray'>\n",
    "balanced accuracy score:  0.6341050254093732"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimators: {'C': 0.05, 'loss': 'squared_hinge', 'penalty': 'l2'}\n",
      "(172,)\n",
      "<class 'numpy.ndarray'>\n",
      "balanced accuracy score:  0.6370808678500987\n"
     ]
    }
   ],
   "source": [
    "#############################################\n",
    "### 2.2 Linear SVC: Hinselmann Prediction ###\n",
    "#############################################\n",
    "\n",
    "# packages\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.pipeline import Pipeline  \n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "SVC = GridSearchCV(estimator=LinearSVC() , param_grid={'penalty':['l1','l2'],'loss':['hinge','squared_hinge'],'C':[0.01,0.05,0.1,0.5,1]},cv=5, n_jobs=-1) # 5-fold CV\n",
    "\n",
    "SVC.fit(X_train, y_train_Hinselmann)\n",
    "\n",
    "print(\"Best estimators: \" + str(SVC.best_params_))\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "y_pred = SVC.predict(X_test)\n",
    "print(y_pred.shape)\n",
    "print(type(y_pred))\n",
    "\n",
    "#>>> y_pred = [0, 1, 0, 0, 0, 1]\n",
    "print(\"balanced accuracy score: \",balanced_accuracy_score(y_test_Hinselmann, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Remember outputs:\n",
    "Best estimators: {'C': 0.5, 'loss': 'squared_hinge', 'penalty': 'l2'}\n",
    "(172,)\n",
    "<class 'numpy.ndarray'>\n",
    "balanced accuracy score:  0.47633136094674555"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimators: {'C': 0.1, 'loss': 'hinge', 'penalty': 'l2'}\n",
      "(172,)\n",
      "<class 'numpy.ndarray'>\n",
      "balanced accuracy score:  0.6760084925690022\n"
     ]
    }
   ],
   "source": [
    "#########################################\n",
    "### 2.3 Linear SVC: Schiller Prediction ###\n",
    "#########################################\n",
    "\n",
    "# packages\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.pipeline import Pipeline  \n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "SVC = GridSearchCV(estimator=LinearSVC() , param_grid={'penalty':['l1','l2'],'loss':['hinge','squared_hinge'],'C':[0.01,0.05,0.1,0.5,1]},cv=5, n_jobs=-1) # 5-fold CV\n",
    "\n",
    "SVC.fit(X_train, y_train_Schiller)\n",
    "\n",
    "print(\"Best estimators: \" + str(SVC.best_params_))\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "y_pred = SVC.predict(X_test)\n",
    "print(y_pred.shape)\n",
    "print(type(y_pred))\n",
    "\n",
    "#>>> y_pred = [0, 1, 0, 0, 0, 1]\n",
    "print(\"balanced accuracy score: \",balanced_accuracy_score(y_test_Schiller, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Remember Outputs:\n",
    "\n",
    "Best estimators: {'C': 0.1, 'loss': 'hinge', 'penalty': 'l2'}\n",
    "(172,)\n",
    "<class 'numpy.ndarray'>\n",
    "balanced accuracy score:  0.6760084925690022"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimators: {'C': 0.01, 'loss': 'squared_hinge', 'penalty': 'l2'}\n",
      "(172,)\n",
      "<class 'numpy.ndarray'>\n",
      "balanced accuracy score:  0.5\n"
     ]
    }
   ],
   "source": [
    "###########################################\n",
    "### 2.4 Linear SVC: Citology Prediction ###\n",
    "###########################################\n",
    "\n",
    "# packages\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.pipeline import Pipeline  \n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "SVC = GridSearchCV(estimator=LinearSVC() , param_grid={'penalty':['l1','l2'],'loss':['hinge','squared_hinge'],'C':[0.01,0.05,0.1,0.5,1]},cv=5, n_jobs=-1) # 5-fold CV\n",
    "\n",
    "SVC.fit(X_train, y_train_Citology)\n",
    "\n",
    "print(\"Best estimators: \" + str(SVC.best_params_))\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "y_pred = SVC.predict(X_test)\n",
    "print(y_pred.shape)\n",
    "print(type(y_pred))\n",
    "\n",
    "#>>> y_pred = [0, 1, 0, 0, 0, 1]\n",
    "print(\"balanced accuracy score: \",balanced_accuracy_score(y_test_Citology, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Remember Outputs:¶\n",
    "\n",
    "Best estimators: {'C': 0.5, 'loss': 'hinge', 'penalty': 'l2'}\n",
    "(172,)\n",
    "<class 'numpy.ndarray'>\n",
    "balanced accuracy score:  0.5432856169052488"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}