Notebooks/Classification of Thyroid Disorders Using Support Vector Machines.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.utils.class_weight import compute_class_weight\n",
    "from sklearn.model_selection import GridSearchCV, train_test_split, cross_val_score\n",
    "from sklearn.metrics import classification_report, accuracy_score, confusion_matrix\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Helper Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_row(row):\n",
    "    row = row.split(\".\")[0]\n",
    "    return row\n",
    "\n",
    "def to_hyperthyroid(row):\n",
    "    if row != \"negative\":\n",
    "        row = \"hyperthyroid\"\n",
    "    return row\n",
    "\n",
    "def to_hypothyroid(row):\n",
    "    if row != \"negative\":\n",
    "        row = \"hypothyroid\"\n",
    "    return row\n",
    "\n",
    "def convert_category(dataframe, column):\n",
    "    \n",
    "    if column == 'Sex':\n",
    "        conditionF = dataframe[column] == 'F' # For sex column\n",
    "        conditionT = dataframe[column] == 'M' # For sex column\n",
    "    else:\n",
    "        conditionF = dataframe[column] == 'f'\n",
    "        conditionT = dataframe[column] == 't'\n",
    "    \n",
    "    dataframe.loc[conditionF, column] = 0\n",
    "    dataframe.loc[conditionT, column] = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Retrieval  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns = [\"Age\", \"Sex\", \"On Thyroxine\", \"Query on Thyroxine\", \n",
    "           \"On Antithyroid Medication\", \"Sick\", \"Pregnant\", \n",
    "           \"Thyroid Surgery\", \"I131 Treatment\", \"Query Hypothyroid\", \n",
    "           \"Query Hyperthyroid\", \"Lithium\", \"Goitre\", \"Tumor\", \n",
    "           \"Hypopituitary\", \"Psych\", \"TSH Measured\", \"TSH\", \"T3 Measured\", \n",
    "           \"T3\", \"TT4 Measured\", \"TT4\", \"T4U Measured\", \"T4U\", \n",
    "           \"FTI Measured\", \"FTI\", \"TBG Measured\", \"TBG\", \"Referral Source\", \"Category\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "hyper_data = pd.read_csv(\"../Datasets/allhyper.csv\", names=columns)\n",
    "hypo_data = pd.read_csv(\"../Datasets/allhypo.csv\", names=columns)\n",
    "sick_data = pd.read_csv(\"../Datasets/sick.csv\", names=columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "hyper_data['Category'] = hyper_data['Category'] \\\n",
    "                            .apply(parse_row) \\\n",
    "                            .apply(to_hyperthyroid)\n",
    "\n",
    "hypo_data['Category'] = hypo_data['Category'] \\\n",
    "                            .apply(parse_row) \\\n",
    "                            .apply(to_hypothyroid)\n",
    "\n",
    "sick_data['Category'] = sick_data['Category'] \\\n",
    "                            .apply(parse_row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Combined Thyroid Data\n",
    "\n",
    "thyroid_frames = [hyper_data, hypo_data, sick_data]\n",
    "thyroid_data = pd.concat(thyroid_frames) \\\n",
    "                 .drop_duplicates() \\\n",
    "                 .drop(['Referral Source', 'TBG', 'TBG Measured'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "classes = thyroid_data['Category'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of samples: 3221\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of samples:\", len(thyroid_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Sex</th>\n",
       "      <th>On Thyroxine</th>\n",
       "      <th>Query on Thyroxine</th>\n",
       "      <th>On Antithyroid Medication</th>\n",
       "      <th>Sick</th>\n",
       "      <th>Pregnant</th>\n",
       "      <th>Thyroid Surgery</th>\n",
       "      <th>I131 Treatment</th>\n",
       "      <th>Query Hypothyroid</th>\n",
       "      <th>...</th>\n",
       "      <th>TSH</th>\n",
       "      <th>T3 Measured</th>\n",
       "      <th>T3</th>\n",
       "      <th>TT4 Measured</th>\n",
       "      <th>TT4</th>\n",
       "      <th>T4U Measured</th>\n",
       "      <th>T4U</th>\n",
       "      <th>FTI Measured</th>\n",
       "      <th>FTI</th>\n",
       "      <th>Category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>41</td>\n",
       "      <td>F</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>...</td>\n",
       "      <td>1.3</td>\n",
       "      <td>t</td>\n",
       "      <td>2.5</td>\n",
       "      <td>t</td>\n",
       "      <td>125</td>\n",
       "      <td>t</td>\n",
       "      <td>1.14</td>\n",
       "      <td>t</td>\n",
       "      <td>109</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>F</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>...</td>\n",
       "      <td>4.1</td>\n",
       "      <td>t</td>\n",
       "      <td>2</td>\n",
       "      <td>t</td>\n",
       "      <td>102</td>\n",
       "      <td>f</td>\n",
       "      <td>?</td>\n",
       "      <td>f</td>\n",
       "      <td>?</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>46</td>\n",
       "      <td>M</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>f</td>\n",
       "      <td>?</td>\n",
       "      <td>t</td>\n",
       "      <td>109</td>\n",
       "      <td>t</td>\n",
       "      <td>0.91</td>\n",
       "      <td>t</td>\n",
       "      <td>120</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>70</td>\n",
       "      <td>F</td>\n",
       "      <td>t</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>...</td>\n",
       "      <td>0.16</td>\n",
       "      <td>t</td>\n",
       "      <td>1.9</td>\n",
       "      <td>t</td>\n",
       "      <td>175</td>\n",
       "      <td>f</td>\n",
       "      <td>?</td>\n",
       "      <td>f</td>\n",
       "      <td>?</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>70</td>\n",
       "      <td>F</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>f</td>\n",
       "      <td>...</td>\n",
       "      <td>0.72</td>\n",
       "      <td>t</td>\n",
       "      <td>1.2</td>\n",
       "      <td>t</td>\n",
       "      <td>61</td>\n",
       "      <td>t</td>\n",
       "      <td>0.87</td>\n",
       "      <td>t</td>\n",
       "      <td>70</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  Age Sex On Thyroxine Query on Thyroxine On Antithyroid Medication Sick  \\\n",
       "0  41   F            f                  f                         f    f   \n",
       "1  23   F            f                  f                         f    f   \n",
       "2  46   M            f                  f                         f    f   \n",
       "3  70   F            t                  f                         f    f   \n",
       "4  70   F            f                  f                         f    f   \n",
       "\n",
       "  Pregnant Thyroid Surgery I131 Treatment Query Hypothyroid  ...   TSH  \\\n",
       "0        f               f              f                 f  ...   1.3   \n",
       "1        f               f              f                 f  ...   4.1   \n",
       "2        f               f              f                 f  ...  0.98   \n",
       "3        f               f              f                 f  ...  0.16   \n",
       "4        f               f              f                 f  ...  0.72   \n",
       "\n",
       "  T3 Measured   T3 TT4 Measured  TT4 T4U Measured   T4U FTI Measured  FTI  \\\n",
       "0           t  2.5            t  125            t  1.14            t  109   \n",
       "1           t    2            t  102            f     ?            f    ?   \n",
       "2           f    ?            t  109            t  0.91            t  120   \n",
       "3           t  1.9            t  175            f     ?            f    ?   \n",
       "4           t  1.2            t   61            t  0.87            t   70   \n",
       "\n",
       "   Category  \n",
       "0  negative  \n",
       "1  negative  \n",
       "2  negative  \n",
       "3  negative  \n",
       "4  negative  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thyroid_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# A quick fix needed\n",
    "thyroid_data.loc[thyroid_data['Age'] == '455', 'Age'] = '45'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Binarize Category Columns\n",
    "binary_cols = ['On Thyroxine', 'Query on Thyroxine', 'Sex',\n",
    "               'On Antithyroid Medication', 'Sick', 'Pregnant', \n",
    "               'Thyroid Surgery', 'I131 Treatment', 'Query Hypothyroid', \n",
    "               'Query Hyperthyroid', 'Lithium', 'Goitre', 'Tumor', \n",
    "               'Hypopituitary', 'Psych', 'TSH Measured', 'T3 Measured', \n",
    "               'TT4 Measured', 'T4U Measured', 'FTI Measured']\n",
    "\n",
    "for col in binary_cols: convert_category(thyroid_data, col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert '?' to np.nan and convert numeric data to numeric dtype\n",
    "for col in thyroid_data.columns: \n",
    "    if col != 'Category':\n",
    "        thyroid_data.loc[thyroid_data[col] == '?', col] = np.nan\n",
    "        thyroid_data[col] = pd.to_numeric(thyroid_data[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "curr_columns = thyroid_data.columns.difference(['Category'])\n",
    "\n",
    "imputer = SimpleImputer(missing_values=np.nan, strategy='median')\n",
    "imputed_data = imputer.fit_transform(thyroid_data.drop('Category', axis=1))\n",
    "imputed_data = pd.DataFrame(imputed_data, columns=curr_columns)\n",
    "\n",
    "thyroid_data = pd.concat([\n",
    "                    imputed_data.reset_index(), \n",
    "                    thyroid_data['Category'].reset_index()], \n",
    "                    axis=1).drop('index', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>FTI</th>\n",
       "      <th>FTI Measured</th>\n",
       "      <th>Goitre</th>\n",
       "      <th>Hypopituitary</th>\n",
       "      <th>I131 Treatment</th>\n",
       "      <th>Lithium</th>\n",
       "      <th>On Antithyroid Medication</th>\n",
       "      <th>On Thyroxine</th>\n",
       "      <th>Pregnant</th>\n",
       "      <th>...</th>\n",
       "      <th>T3 Measured</th>\n",
       "      <th>T4U</th>\n",
       "      <th>T4U Measured</th>\n",
       "      <th>TSH</th>\n",
       "      <th>TSH Measured</th>\n",
       "      <th>TT4</th>\n",
       "      <th>TT4 Measured</th>\n",
       "      <th>Thyroid Surgery</th>\n",
       "      <th>Tumor</th>\n",
       "      <th>Category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>41.0</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",
       "      <td>...</td>\n",
       "      <td>1.30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>23.0</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",
       "      <td>...</td>\n",
       "      <td>4.10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>102.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>46.0</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",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.91</td>\n",
       "      <td>1.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>70.0</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",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.16</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1.0</td>\n",
       "      <td>175.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>70.0</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",
       "      <td>...</td>\n",
       "      <td>0.72</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.87</td>\n",
       "      <td>1.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age  FTI  FTI Measured  Goitre  Hypopituitary  I131 Treatment  Lithium  \\\n",
       "0  41.0  0.0           0.0     0.0            0.0             0.0      0.0   \n",
       "1  23.0  0.0           0.0     0.0            0.0             0.0      0.0   \n",
       "2  46.0  1.0           0.0     0.0            0.0             0.0      0.0   \n",
       "3  70.0  0.0           1.0     0.0            0.0             0.0      0.0   \n",
       "4  70.0  0.0           0.0     0.0            0.0             0.0      0.0   \n",
       "\n",
       "   On Antithyroid Medication  On Thyroxine  Pregnant  ...  T3 Measured  T4U  \\\n",
       "0                        0.0           0.0       0.0  ...         1.30  1.0   \n",
       "1                        0.0           0.0       0.0  ...         4.10  1.0   \n",
       "2                        0.0           0.0       0.0  ...         0.98  0.0   \n",
       "3                        0.0           0.0       0.0  ...         0.16  1.0   \n",
       "4                        0.0           0.0       0.0  ...         0.72  1.0   \n",
       "\n",
       "   T4U Measured  TSH  TSH Measured  TT4  TT4 Measured  Thyroid Surgery  Tumor  \\\n",
       "0           2.5  1.0         125.0  1.0          1.14              1.0  109.0   \n",
       "1           2.0  1.0         102.0  0.0          0.97              0.0  106.0   \n",
       "2           1.9  1.0         109.0  1.0          0.91              1.0  120.0   \n",
       "3           1.9  1.0         175.0  0.0          0.97              0.0  106.0   \n",
       "4           1.2  1.0          61.0  1.0          0.87              1.0   70.0   \n",
       "\n",
       "   Category  \n",
       "0  negative  \n",
       "1  negative  \n",
       "2  negative  \n",
       "3  negative  \n",
       "4  negative  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thyroid_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Split Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = thyroid_data.drop('Category', axis=1)\n",
    "y = thyroid_data['Category']\n",
    "\n",
    "col_names = X.columns\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, stratify=y)\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "X_train = pd.DataFrame(scaler.fit_transform(X_train))\n",
    "X_test = pd.DataFrame(scaler.transform(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize Data with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_pca_data(X, y):    \n",
    "    \n",
    "    LE = LabelEncoder()\n",
    "    y_encoded = LE.fit_transform(y)\n",
    "    \n",
    "    pca = PCA(n_components=2)\n",
    "    pca.fit(X)\n",
    "    pca_X = pca.transform(X)\n",
    "\n",
    "    x_axis = pca_X[:,0]\n",
    "    y_axis = pca_X[:,1]\n",
    "\n",
    "    fig = plt.figure(figsize=(10,10))\n",
    "    ax = fig.add_subplot(111)\n",
    "    \n",
    "    for label in np.unique(y):\n",
    "        ax.scatter(pca_X[y==label, 0],\n",
    "                   pca_X[y==label, 1], \n",
    "                   label=label,\n",
    "                   s=100,\n",
    "                   edgecolor='k')\n",
    "        \n",
    "\n",
    "    ax.legend()\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "\n",
    "plot_pca_data(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def search_for_parameters(estimator, X_data, y_data, grid_parameters, scoring='accuracy', cv=3):\n",
    "    \n",
    "    classes = y_data.unique()\n",
    "    \n",
    "    # Execute Grid Search\n",
    "    grid_clf = GridSearchCV(estimator=estimator, scoring=scoring,\n",
    "                            param_grid=grid_parameters, iid=False,\n",
    "                            cv=cv,n_jobs=-1)\n",
    "    \n",
    "    grid_clf.fit(X_data, y_data)\n",
    "\n",
    "    print(\"The best parameters are: \", grid_clf.best_params_)\n",
    "\n",
    "    return grid_clf.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best parameters are:  {'C': 8, 'degree': 5, 'gamma': 'scale', 'kernel': 'poly'}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SVC(C=8, cache_size=200, class_weight='balanced', coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=5, gamma='scale', kernel='poly',\n",
       "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "    tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_clf = SVC(class_weight='balanced')\n",
    "\n",
    "svm_params_list = {'C':[1,2,4,8], \n",
    "                   'kernel':['poly', 'rbf', 'sigmoid'], \n",
    "                   'degree':[3,4,5], \n",
    "                   'gamma':['auto','scale']}\n",
    "\n",
    "svm_parameters = search_for_parameters(estimator=svm_clf, \n",
    "                                X_data=X_train, \n",
    "                                y_data=y_train, cv=5,\n",
    "                                grid_parameters=svm_params_list)\n",
    "\n",
    "svm_clf.set_params(**svm_parameters)\n",
    "\n",
    "svm_clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluate Parameters "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Scoring Evaluation Results\n",
      "The mean score and the confidence interval of the score estimate are:\n",
      "Accuracy: 0.69 (+/- 0.03)\n"
     ]
    }
   ],
   "source": [
    "def cross_validate(estimator, X_data, y_data, scoring='accuracy', cv=3, Z=2):\n",
    "    \n",
    "    classes = y_data.unique()\n",
    "    \n",
    "    # Execute Cross Validation\n",
    "    scores = cross_val_score(estimator=estimator, X=X_data, y=y_data, cv=cv, scoring=scoring)\n",
    "\n",
    "    print(\"Model Scoring Evaluation Results\")\n",
    "    print(\"The mean score and the confidence interval of the score estimate are:\")\n",
    "    print(\"Accuracy: %0.2f (+/- %0.2f)\" % (scores.mean(), scores.std() * Z))\n",
    "\n",
    "cross_validate(estimator=svm_clf, X_data=X_train, y_data=y_train, cv=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predict on Test Set "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6935483870967742\n",
      "**************************************************\n",
      "\n",
      "\t\t\tClassification Report\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "hyperthyroid       0.21      0.68      0.32        19\n",
      " hypothyroid       0.25      0.64      0.36        55\n",
      "    negative       0.95      0.70      0.80       689\n",
      "        sick       0.30      0.67      0.42        43\n",
      "\n",
      "    accuracy                           0.69       806\n",
      "   macro avg       0.43      0.67      0.48       806\n",
      "weighted avg       0.85      0.69      0.74       806\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y_true, y_pred = y_test, svm_clf.predict(X_test)\n",
    "\n",
    "print(\"Accuracy:\",accuracy_score(y_test, y_pred))\n",
    "print(\"*\"*50)\n",
    "print()\n",
    "print(\"\\t\\t\\tClassification Report\")\n",
    "print()\n",
    "print(classification_report(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[482  46 100  61]\n",
      " [  5  13   0   1]\n",
      " [ 15   0  35   5]\n",
      " [  7   3   4  29]]\n"
     ]
    }
   ],
   "source": [
    "# Confusion Matrix\n",
    "cm = confusion_matrix(y_true, y_pred, labels=classes)\n",
    "print(cm)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}