Code/Model1_SVM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import missingno as msno\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = 'all'\n",
    "\n",
    "seed = 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read the CSV and Perform Basic Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>koi_disposition</th>\n",
       "      <th>koi_fpflag_nt</th>\n",
       "      <th>koi_fpflag_ss</th>\n",
       "      <th>koi_fpflag_co</th>\n",
       "      <th>koi_fpflag_ec</th>\n",
       "      <th>koi_period</th>\n",
       "      <th>koi_period_err1</th>\n",
       "      <th>koi_period_err2</th>\n",
       "      <th>koi_time0bk</th>\n",
       "      <th>koi_time0bk_err1</th>\n",
       "      <th>...</th>\n",
       "      <th>koi_steff_err2</th>\n",
       "      <th>koi_slogg</th>\n",
       "      <th>koi_slogg_err1</th>\n",
       "      <th>koi_slogg_err2</th>\n",
       "      <th>koi_srad</th>\n",
       "      <th>koi_srad_err1</th>\n",
       "      <th>koi_srad_err2</th>\n",
       "      <th>ra</th>\n",
       "      <th>dec</th>\n",
       "      <th>koi_kepmag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CONFIRMED</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>54.418383</td>\n",
       "      <td>2.479000e-04</td>\n",
       "      <td>-2.479000e-04</td>\n",
       "      <td>162.513840</td>\n",
       "      <td>0.003520</td>\n",
       "      <td>...</td>\n",
       "      <td>-81</td>\n",
       "      <td>4.467</td>\n",
       "      <td>0.064</td>\n",
       "      <td>-0.096</td>\n",
       "      <td>0.927</td>\n",
       "      <td>0.105</td>\n",
       "      <td>-0.061</td>\n",
       "      <td>291.93423</td>\n",
       "      <td>48.141651</td>\n",
       "      <td>15.347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>FALSE POSITIVE</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19.899140</td>\n",
       "      <td>1.490000e-05</td>\n",
       "      <td>-1.490000e-05</td>\n",
       "      <td>175.850252</td>\n",
       "      <td>0.000581</td>\n",
       "      <td>...</td>\n",
       "      <td>-176</td>\n",
       "      <td>4.544</td>\n",
       "      <td>0.044</td>\n",
       "      <td>-0.176</td>\n",
       "      <td>0.868</td>\n",
       "      <td>0.233</td>\n",
       "      <td>-0.078</td>\n",
       "      <td>297.00482</td>\n",
       "      <td>48.134129</td>\n",
       "      <td>15.436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>FALSE POSITIVE</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.736952</td>\n",
       "      <td>2.630000e-07</td>\n",
       "      <td>-2.630000e-07</td>\n",
       "      <td>170.307565</td>\n",
       "      <td>0.000115</td>\n",
       "      <td>...</td>\n",
       "      <td>-174</td>\n",
       "      <td>4.564</td>\n",
       "      <td>0.053</td>\n",
       "      <td>-0.168</td>\n",
       "      <td>0.791</td>\n",
       "      <td>0.201</td>\n",
       "      <td>-0.067</td>\n",
       "      <td>285.53461</td>\n",
       "      <td>48.285210</td>\n",
       "      <td>15.597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CONFIRMED</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.525592</td>\n",
       "      <td>3.760000e-06</td>\n",
       "      <td>-3.760000e-06</td>\n",
       "      <td>171.595550</td>\n",
       "      <td>0.001130</td>\n",
       "      <td>...</td>\n",
       "      <td>-211</td>\n",
       "      <td>4.438</td>\n",
       "      <td>0.070</td>\n",
       "      <td>-0.210</td>\n",
       "      <td>1.046</td>\n",
       "      <td>0.334</td>\n",
       "      <td>-0.133</td>\n",
       "      <td>288.75488</td>\n",
       "      <td>48.226200</td>\n",
       "      <td>15.509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CONFIRMED</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.134435</td>\n",
       "      <td>1.050000e-05</td>\n",
       "      <td>-1.050000e-05</td>\n",
       "      <td>172.979370</td>\n",
       "      <td>0.001900</td>\n",
       "      <td>...</td>\n",
       "      <td>-232</td>\n",
       "      <td>4.486</td>\n",
       "      <td>0.054</td>\n",
       "      <td>-0.229</td>\n",
       "      <td>0.972</td>\n",
       "      <td>0.315</td>\n",
       "      <td>-0.105</td>\n",
       "      <td>296.28613</td>\n",
       "      <td>48.224670</td>\n",
       "      <td>15.714</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  koi_disposition  koi_fpflag_nt  koi_fpflag_ss  koi_fpflag_co  koi_fpflag_ec  \\\n",
       "0       CONFIRMED              0              0              0              0   \n",
       "1  FALSE POSITIVE              0              1              0              0   \n",
       "2  FALSE POSITIVE              0              1              0              0   \n",
       "3       CONFIRMED              0              0              0              0   \n",
       "4       CONFIRMED              0              0              0              0   \n",
       "\n",
       "   koi_period  koi_period_err1  koi_period_err2  koi_time0bk  \\\n",
       "0   54.418383     2.479000e-04    -2.479000e-04   162.513840   \n",
       "1   19.899140     1.490000e-05    -1.490000e-05   175.850252   \n",
       "2    1.736952     2.630000e-07    -2.630000e-07   170.307565   \n",
       "3    2.525592     3.760000e-06    -3.760000e-06   171.595550   \n",
       "4    4.134435     1.050000e-05    -1.050000e-05   172.979370   \n",
       "\n",
       "   koi_time0bk_err1  ...  koi_steff_err2  koi_slogg  koi_slogg_err1  \\\n",
       "0          0.003520  ...             -81      4.467           0.064   \n",
       "1          0.000581  ...            -176      4.544           0.044   \n",
       "2          0.000115  ...            -174      4.564           0.053   \n",
       "3          0.001130  ...            -211      4.438           0.070   \n",
       "4          0.001900  ...            -232      4.486           0.054   \n",
       "\n",
       "   koi_slogg_err2  koi_srad  koi_srad_err1  koi_srad_err2         ra  \\\n",
       "0          -0.096     0.927          0.105         -0.061  291.93423   \n",
       "1          -0.176     0.868          0.233         -0.078  297.00482   \n",
       "2          -0.168     0.791          0.201         -0.067  285.53461   \n",
       "3          -0.210     1.046          0.334         -0.133  288.75488   \n",
       "4          -0.229     0.972          0.315         -0.105  296.28613   \n",
       "\n",
       "         dec  koi_kepmag  \n",
       "0  48.141651      15.347  \n",
       "1  48.134129      15.436  \n",
       "2  48.285210      15.597  \n",
       "3  48.226200      15.509  \n",
       "4  48.224670      15.714  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(6991, 41)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"../Data/exoplanet_data.csv\")\n",
    "# Drop the null columns where all values are null\n",
    "df = df.dropna(axis='columns', how='all') #No null columns\n",
    "# Drop the null rows\n",
    "df = df.dropna() #No missing values\n",
    "df.head()\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data attributes and desctiption\n",
    "\n",
    "From [exoplanet archive](https://exoplanetarchive.ipac.caltech.edu/docs/API_kepcandidate_columns.html#pdisposition), the following descriptions are retrieved. `Kepler Object of Interest` is acronymed as `KOI`.\n",
    "\n",
    "1. koi_disposition : The category of this KOI from the Exoplanet Archive. Current values are CANDIDATE, FALSE POSITIVE, NOT DISPOSITIONED or CONFIRMED. (THIS is the target value)\n",
    "1. koi_fpflag_nt : Not Transit-Like Flag (A KOI whose light curve is not consistent with that of a transiting planet. This includes, but is not limited to, instrumental artifacts, non-eclipsing variable stars, and spurious (very low SNR) detections.) (Boolean)\n",
    "1. koi_fpflag_ss : Stellar Eclipse Flag. A KOI that is observed to have a significant secondary event, transit shape, or out-of-eclipse variability, which indicates that the transit-like event is most likely caused by an eclipsing binary. However, self-luminous, hot Jupiters with a visible secondary eclipse will also have this flag set, but with a disposition of PC. (Boolean)\n",
    "1. koi_fpflag_co : Centroid Offset Flag. The source of the signal is from a nearby star, as inferred by measuring the centroid location of the image both in and out of transit, or by the strength of the transit signal in the target's outer (halo) pixels as compared to the transit signal from the pixels in the optimal (or core) aperture. (Boolean)\n",
    "1. koi_fpflag_ec : Ephemeris Match Indicates Contamination Flag. The KOI shares the same period and epoch as another object and is judged to be the result of flux contamination in the aperture or electronic crosstalk. (Boolean)\n",
    "1. koi_period : Orbital Period (days). The interval between consecutive planetary transits. (float)\n",
    "1. koi_period_err1: koi period positive error bar (float)\n",
    "1. koi_period_err2: koi period negative error bar (float)\n",
    "1. koi_time0bk: Transit Epoch (BJD - 2,454,833.0). The time corresponding to the center of the first detected transit in Barycentric Julian Day (BJD) minus a constant offset of 2,454,833.0 days. The offset corresponds to 12:00 on Jan 1, 2009 UTC. (float)\n",
    "1. koi_time0bk_err1: koi time0bk positive error bar (float)\n",
    "1. koi_time0bk_err2: koi time0bk negative error bar (float)\n",
    "1. koi_impact: \tImpact Parameter. The sky-projected distance between the center of the stellar disc and the center of the planet disc at conjunction, normalized by the stellar radius. (float)\n",
    "1. koi_impact_err1: koi impact positive error bar (float)\n",
    "1. koi_impact_err2: koi impact negative error bar (float)\n",
    "1. koi_duration: Transit Duration (hours). The duration of the observed transits. Duration is measured from first contact between the planet and star until last contact. Contact times are typically computed from a best-fit model produced by a Mandel-Agol (2002) model fit to a multi-quarter Kepler light curve, assuming a linear orbital ephemeris. (float)\n",
    "1. koi_duration_err1: koi duration positive error bar (float)\n",
    "1. koi_duration_err2: koi duration negative error bar (float)\n",
    "1. koi_depth: Transit Depth (parts per million). The fraction of stellar flux lost at the minimum of the planetary transit. Transit depths are typically computed from a best-fit model produced by a Mandel-Agol (2002) model fit to a multi-quarter Kepler light curve, assuming a linear orbital ephemeris. (float)\n",
    "1. koi_depth_err1: koi depth positive error bar (float)\n",
    "1. koi_depth_err2: koi depth negative error bar (float)\n",
    "1. koi_prad: Planetary Radius (Earth radii). The radius of the planet. Planetary radius is the product of the planet star radius ratio and the stellar radius (float)\n",
    "1. koi_prad_err1: koi prad positive error bar (float)\n",
    "1. koi_prad_err2: koi prad negative error bar (float)\n",
    "1. koi_teq: Equilibrium Temperature (Kelvin). Approximation for the temperature of the planet. The calculation of equilibrium temperature assumes a) thermodynamic equilibrium between the incident stellar flux and the radiated heat from the planet, b) a Bond albedo (the fraction of total power incident upon the planet scattered back into space) of 0.3, c) the planet and star are blackbodies, and d) the heat is evenly distributed between the day and night sides of the planet. (int)\n",
    "1. koi_insol: Insolation Flux [Earth flux]: Insolation flux is another way to give the equilibrium temperature. It depends on the stellar parameters (specifically the stellar radius and temperature), and on the semi-major axis of the planet. It's given in units relative to those measured for the Earth from the Sun. (float)\n",
    "1. koi_insol_err1: koi insol positive error bar (float)\n",
    "1. koi_insol_err2: koi insol negative error bar (float)\n",
    "1. koi_model_snr: Transit Signal-to-Noise. Transit depth normalized by the mean uncertainty in the flux during the transits. (float)\n",
    "1. koi_tce_plnt_num: TCE Planet Number. TCE Planet Number federated to the KOI. (int)\n",
    "1. koi_steff: Stellar Effective Temperature (Kelvin). The photospheric temperature of the star. (int)\n",
    "1. koi_steff_err1: koi steff positive error bar (int)\n",
    "1. koi_steff_err2: koi steff negative error bar (int)\n",
    "1. koi_slogg: Stellar Surface Gravity (log10(cm s-2)). The base-10 logarithm of the acceleration due to gravity at the surface of the star. (float)\n",
    "1. koi_slogg_err1: koi slogg positive error bar (float)\n",
    "1. koi_slogg_err2: koi slogg negative error bar (float)\n",
    "1. koi_srad: Stellar Radius (solar radii). The photospheric radius of the star (float)\n",
    "1. koi_srad_err1: koi srad positive error bar (float)\n",
    "1. koi_srad_err2: koi srad negative error bar (float)\n",
    "1. ra: RA (deg): KIC Right Ascension (float)\n",
    "1. dec: Dec (deg). KIC Declination (float)\n",
    "1. koi_kepmag: Kepler-band (mag). Kepler-band (mag) (float)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Processing and Model Pipeline\n",
    "\n",
    "![pipeline](../Images/pipeline.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[df.columns.difference(['koi_disposition'])]\n",
    "Y = df['koi_disposition']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, random_state=seed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Selection\n",
    "\n",
    "https://machinelearningmastery.com/rfe-feature-selection-in-python/#:~:text=RFE%20is%20a%20wrapper%2Dtype%20feature%20selection%20algorithm.&text=This%20is%20achieved%20by%20fitting,specified%20number%20of%20features%20remains.\n",
    "\n",
    "**Approach**\n",
    "\n",
    "- Use RFE wrapper with Random Forest Estimator for feature selection.\n",
    "- Use repeated stratified k-fold cross-validation, with 3 repeats and 5 folds.\n",
    "- Use SVM classifier to train the data, and validate\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold, cross_val_score\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "num_features = [1]+list(range(5,len(X_train.columns)+5,5))\n",
    "scores = []\n",
    "for no in num_features:\n",
    "    # create pipeline\n",
    "    scaler = MinMaxScaler()\n",
    "    rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=no, verbose=5)\n",
    "    model = SVC(kernel='rbf', C=100)\n",
    "    pipeline = Pipeline(steps=[('scale', scaler), ('feature_selection',rfe),('model',model)])\n",
    "    # evaluate model\n",
    "    cv = RepeatedStratifiedKFold(n_splits=5, n_repeats=3)\n",
    "    n_scores = cross_val_score(pipeline, X_train, y_train, scoring='accuracy', cv=cv, n_jobs=-1, error_score='raise')\n",
    "    # Append the scores for the \"no\" of features\n",
    "    scores.append(n_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAb90lEQVR4nO3dcZCU9Z3n8feHESFZNzojk60IEkgKs6NcNJc+k1M2G7ILoncl2WhtMXuXkisSjqqF7JlcTgyWGqzxcn9skirLyiwJnlbudojrRWRzKQkLWLvsJRsaBQTmkAnZrLNYoZVJcomAw8z3/uhn9JmhZ+YZ6elunvm8qrqm+/c8T/d3HphPP/17nv79FBGYmVl+Tat3AWZmNrkc9GZmOeegNzPLOQe9mVnOOejNzHLuknoXMNKsWbNi3rx59S7DzOyism/fvlcjorXSsoYL+nnz5lEsFutdhpnZRUXSz0Zb5q4bM7Occ9CbmeWcg97MLOcc9GZmOeegNzPLOQe9mVnOOejNLlJdXV0sXLiQpqYmFi5cSFdXV71LsgaVKeglLZN0VFKPpPUVlr9X0k5JByU9J2lOatldko4lt7uqWbzZVNXV1cWGDRt45JFHOHPmDI888ggbNmxw2FtFGm88eklNwEvAEqAX2Au0R8SR1Dp/BXwvIp6Q9AngP0TEpyW1AEWgAASwD/hwRPSN9nqFQiH8hSmzsS1cuJBHHnmExYsXv9m2e/du1q1bx6FDh+pYWeOQlGm9vMzJIWlfRBQqLctyRH8j0BMRxyPiDWALsHzEOtcCO5P7u1PLbwF2RMSpJNx3AMsm+guY1VujdZN0d3ezaNGiYW2LFi2iu7u7ThU1nogYdqvUVuuQl5TpVm1Zgn428HLqcW/SlnYAuCO5/0fAb0u6MuO2SFotqSipWCqVstZuVhON2E3S1tbGnj17hrXt2bOHtra2OlVkWYz2RjPZbz5Zgr7S28vISv4z8PuSXgB+H/hn4FzGbYmITRFRiIhCa2vFMXnM6qajo4PNmzezePFipk+fzuLFi9m8eTMdHR11q2nDhg2sWrWK3bt309/fz+7du1m1ahUbNmyoW03WuLIMatYLXJ16PAc4kV4hIk4AnwKQdBlwR0T8UlIv8PER2z53AfWa1VwjdpO0t7cDsG7dOrq7u2lra6Ojo+PNdrO0LEf0e4EFkuZLuhRYAWxLryBplqSh57oXeCy5vx1YKqlZUjOwNGkzu2g0ajdJe3s7hw4dYmBggEOHDjnkbVTjBn1EnAPWUg7obuDJiDgsaaOk25PVPg4clfQS8DtAR7LtKeAhym8We4GNSZvZRaORukmynsybjBN61ajL6mPcyytrzZdXTk2NfilcV1cXHR0db3aTbNiwoWGOoCU13CWCrim7atV1oZdXmk26el2NkFV7ezu7f7ybT//vT/Pc3ucaJuRLr5eYv34+r55+td6lvKkeNbW0tIz5SWL6FdOZv34+06+YPu6njpaWlprUNJG6LrQmB73VXCP9AYxl5Gtdv+Z6iq8U+eB//GDNuiPG21fXr7med17zzvNqquV+GqnzYCfvvOaddB7orNlr9vX1VTxYGLrd/+z9XPa7l/HAsw+MuV5E0Nc36vc5q1rTROq60Joc9FZzjfQHkDYyVNMuufwSmn+vGU0Tzb/XzCWXv3XB2mSG6lj76uRvTnLV0qvQNHHV0qsovV6qy34a+Qa95cUtaJroerGrrm/UQ0qvl3im5xmCYGvP1ob59FPLuhpuzlizkX8Aa65fw6x3zJr01z31uQHgXRWXPXRlM09fIvqBGZeIBx6azX2vjRaeA5NV4jCdBzsZjEEABmOQzgOd3PfR+yb9dYfefCp56EcP8fSxp+kf7GfGzBk88OwDY9ZUrU9E8cC74MHLKy7rvLKZwcsug2lisP8Mnd8qjPFvlzxXDdTy388nY632RvmDHPLQlc08fdll9E8T0weDT/3612P+YfLgL6tS1mgnxUqvl7j1u7dyduDsm20zmmbw7B3PVnwDqupJv1H2ValpGrfOuYqz0976UD5jcJBne08wa2BwlOeqzn6qak1Vqqta/3ZjPdeEjfH/fDL+/cY6Geugn8LqdSXJWH9I9QzV0Y4u3/Pp99D8sWamTX/rj3Kwf5C+v+3jlW+/ct76zc3NnDpVnauI81xTNesa7f9B+hPGkOnTpvOpBZ8a9ei5ilfBZPrkk6WuLDX5qhs7T73Hbxnr5OLpM6eHrXv6zOlRTzY2NzdXrabR+rdvuvOmYeEFMG36NG6686aK61crUPNeU7XrqvT/44kfPDEsTAH6B/t5/AePj/p/sJr/p0Z7jYnWdaE1+Yh+imrUYW7v3HYnR/uOntf+geYP8NTtT9WhouGq2i1TJa4pm0asCWpzHb2DfopqamrizJkzTJ8+/c22/v5+Zs6cycBAbU4mjqcR/zBd0+g1ZFHPOhthP1VSi6B3180U1ajjtzSSSh+hK7Xb6N05I29WHw76KaqRxm8ZaWiSD6Cuk3w4vCwv3HUzhTXS+C0Xw0f/RjCRTxDeV433f2oya3LXjVXUSMPcDh0dX3fddezatWvYEfOuXbu47rrrpnxwQfZPGd5XjfmJrF41+YjeGsrFcJLYrBH5iN6GGWvskXqfZPRJYrPqyxT0kpZJOiqpR9L6CsvnStot6QVJByXdlrTPk3Ra0v7kVrvh7GxU6Y+IJ39zkvnr51ccEKseGvkksdnFatxBzSQ1AY8CSyjPH7tX0raIOJJa7T7KM099Q9K1wPeBecmyn0TEDdUt26olPaRsLQbEGo/nQjWrvixH9DcCPRFxPCLeALYAy0esE7w17N/ljJg83OrnYhtSFhrrJLFZHmQZpng28HLqcS/wkRHrPAj8QNI64LeAP0wtmy/pBeBXwH0R8Xdvv1ybqOoNvQu1Gn7XzKorS9BXOis3sgO3HXg8Iv5c0r8Gvi1pIfAKMDciXpP0YWCrpOsi4lfDXkBaDawGmDt37oR/CRudvvyrUYdvfea7t9KfjBLZP01sbZ7Fms8Uxx6+9cHJrNbMJkOWrpte4OrU4zmc3zWzCngSICJ+CMwEZkXE2Yh4LWnfB/wEuGbkC0TEpogoREShtbV14r+FTVh60oMhQ5MfmFm+ZAn6vcACSfMlXQqsALaNWOefgD8AkNRGOehLklqTk7lIeh+wADhereItm0YcvtXMamfcrpuIOCdpLbAdaAIei4jDkjYCxYjYBnwB+Kakuyl366yMiJD0MWCjpHOUO3jXRET1BqC2cWW5TPK80fPun8SCzKzm/M1Ya9jhW80sO38z1sxsCnPQm5nlXJbLKy1nKo1jU6nN3Tlm+eCgn4Ic4GZTi7tuzMxyzkFvZpZzDnozs5xz0JuZ5ZyD3sws5xz0ZmY556A3M8s5B72ZWc456M3Mcs5Bb2aWcw56M7Occ9CbmeVcpqCXtEzSUUk9ktZXWD5X0m5JL0g6KOm21LJ7k+2OSrqlmsWbmdn4xh29Mpnz9VFgCeWJwvdK2hYRR1Kr3Qc8GRHfkHQt8H1gXnJ/BXAdcBXwN5KuiYiBav8iZmZWWZYj+huBnog4HhFvAFuA5SPWCeBdyf3LgRPJ/eXAlog4GxE/BXqS55syRptoe+TNzGyyZAn62cDLqce9SVvag8C/l9RL+Wh+3QS2RdJqSUVJxVKplLH0i0NEDLtVavP48GY2mbIEfaXDzZHJ1A48HhFzgNuAb0ualnFbImJTRBQiotDa2pqhJDMzyyrLDFO9wNWpx3N4q2tmyCpgGUBE/FDSTGBWxm3NzGwSZTmi3wsskDRf0qWUT65uG7HOPwF/ACCpDZgJlJL1VkiaIWk+sAD4cbWKNzOz8Y17RB8R5yStBbYDTcBjEXFY0kagGBHbgC8A35R0N+WumZVR7ng+LOlJ4AhwDvhTX3FjZlZbarQTgYVCIYrFYr3LmDSSfPLVzKpO0r6IKFRa5m/GmpnlnIPezCznHPRmZjnnoDczyzkHfRW1tLRkGuogy5AILS0tdf5tzCwvsnxhyjLq6+ur2hU1Hv/GzKrFR/RmZjnnoDczyzkHvZlZzjnozcxyzkFvZpZzDnozs5xz0JuZ5ZyD3sws5xz0ZmY556A3M8u5TEEvaZmko5J6JK2vsPxrkvYnt5ck/SK1bCC1bOQUhGZmNsnGHetGUhPwKLCE8mTfeyVti4gjQ+tExN2p9dcBH0o9xemIuKF6JZuZ2URkOaK/EeiJiOMR8QawBVg+xvrtQFc1ijMzswuXJehnAy+nHvcmbeeR9F5gPrAr1TxTUlHSjyR9cpTtVifrFEulUsbSzcwsiyxBX2m83NHG4l0BPBURA6m2ucmEtX8CfF3S+897sohNEVGIiEJra2uGkszMLKssQd8LXJ16PAc4Mcq6KxjRbRMRJ5Kfx4HnGN5/b2ZmkyxL0O8FFkiaL+lSymF+3tUzkj4ANAM/TLU1S5qR3J8F3AwcGbntVFF6vcTKZ1fy6ulX612KmU0h4wZ9RJwD1gLbgW7gyYg4LGmjpNtTq7YDW2L4FEttQFHSAWA38JX01TpTTefBTp7/+fN0HuisdylmNoWoWlPfVUuhUIhisVjvMt4WSaNOJVh6vcSt372VswNnmdE0g2fveJZZ75j1tp7LzGwkSfuS86Hn8Tdja6TzYCeDMQjAYAz6qN7MasZBXwOl10s80/MM/YP9APQP9rO1Z6v76s2sJhz0NZA+mh/io3ozqxUHfQ0cOHngzaP5If2D/ew/ub9OFZnZVDLuWDd24Z66/al6l2BmU5iP6M3Mcs5Bb2aWcw56M7Occx99FcUD74IHL6/ec5mZVYGDvor05V9V7duskogHq/JUZjbFuevGzCznHPRmZjnnoDczyzn30VeZVGlCrolrbm6uyvOYmTnoqyjLiVgPP2xmtZap60bSMklHJfVIWl9h+dck7U9uL0n6RWrZXZKOJbe7qlm8mZmNb9wjeklNwKPAEsrzx+6VtC09U1RE3J1afx3JvLCSWoAHgALlCcX3Jdv2VfW3MDOzUWU5or8R6ImI4xHxBrAFWD7G+u28NUH4LcCOiDiVhPsOYNmFFGxmZhOTJehnAy+nHvcmbeeR9F5gPrBrIttKWi2pKKlYKpWy1G1mZhllCfpKl5GMdjZxBfBURAxMZNuI2BQRhYgotLa2ZijJzMyyyhL0vcDVqcdzgBOjrLuCt7ptJrqtmZlNgixBvxdYIGm+pEsph/m2kStJ+gDQDPww1bwdWCqpWVIzsDRpMzOzGhn3qpuIOCdpLeWAbgIei4jDkjYCxYgYCv12YEukLhKPiFOSHqL8ZgGwMSJOVfdXMDOzsajRvrxTKBSiWCzWu4yqyfpN2Ub7dzCzi4ukfRFRqLTM34ydZA5wM6s3D2pmZpZzDnozs5xz0JuZ5ZyD3sws5xz0ZmY556A3M8s5B72ZWc456M3Mcs5Bb2aWcw56M7Occ9CbmeWcg97MLOcc9GZmOeegNzPLOQe9mVnOZQp6ScskHZXUI2n9KOv8saQjkg5L+stU+4Ck/cntvCkIzcxsco078YikJuBRYAnlyb73StoWEUdS6ywA7gVujog+Se9OPcXpiLihynWbmVlGWY7obwR6IuJ4RLwBbAGWj1jns8CjEdEHEBEnq1ummZm9XVmCfjbwcupxb9KWdg1wjaS/l/QjSctSy2ZKKibtn6z0ApJWJ+sUS6XShH4BMzMbW5Y5YyvNbj1yItRLgAXAx4E5wN9JWhgRvwDmRsQJSe8Ddkl6MSJ+MuzJIjYBm6A8OfgEfwczMxtDliP6XuDq1OM5wIkK6zwTEf0R8VPgKOXgJyJOJD+PA88BH7rAms3MbAKyBP1eYIGk+ZIuBVYAI6+e2QosBpA0i3JXznFJzZJmpNpvBo5gZmY1M27XTUSck7QW2A40AY9FxGFJG4FiRGxLli2VdAQYAL4YEa9Jugn4C0mDlN9UvpK+WsfMzCafIhqrS7xQKESxWKx3GWZmFxVJ+yKiUGmZvxlrZpZzDnozs5xz0JuZ5ZyD3sws5xz0ZmY556A3M8s5B72ZWc456M3Mcs5Bb2aWcw56M7Occ9CbmeWcg97MLOcc9GZmOeegNzPLOQe9mVnOZQp6ScskHZXUI2n9KOv8saQjkg5L+stU+12SjiW3u6pVuJmZZTPuDFOSmoBHgSWU54bdK2lbeqYoSQuAe4GbI6JP0ruT9hbgAaBAeULxfcm2fdX/VczMrJIsR/Q3Aj0RcTwi3gC2AMtHrPNZ4NGhAI+Ik0n7LcCOiDiVLNsBLKtO6WZmlkWWoJ8NvJx63Ju0pV0DXCPp7yX9SNKyCWyLpNWSipKKpVIpe/VmZjauLEGvCm0jJ5q9BFgAfBxoB74l6YqM2xIRmyKiEBGF1tbWDCWZmVlWWYK+F7g69XgOcKLCOs9ERH9E/BQ4Sjn4s2xrZmaTKEvQ7wUWSJov6VJgBbBtxDpbgcUAkmZR7so5DmwHlkpqltQMLE3azMysRsa96iYizklaSzmgm4DHIuKwpI1AMSK28VagHwEGgC9GxGsAkh6i/GYBsDEiTk3GL2JmZpUp4rwu87oqFApRLBbrXYaZ2UVF0r6IKFRa5m/GmpnlnIPezCznHPRmZjnnoDczyzkHvZlZzjnozcxyzkFvZpZzDnozs5xz0JuZ5ZyD3sws5xz0ZmY556A3M8s5B72ZWc7lMujXrVvHzJkzkcTMmTNZt25dvUsyM6ub3AX9unXr6Ozs5OGHH+Y3v/kNDz/8MJ2dnQ57M5uycjce/cyZM3n44Yf5/Oc//2bbV7/6Vb70pS9x5syZapRoZtZwLng8eknLJB2V1CNpfYXlKyWVJO1Pbp9JLRtItY+cgrDqzp49y5o1a4a1rVmzhrNnz072S5uZNaRxg15SE/AocCtwLdAu6doKq34nIm5Ibt9KtZ9Otd9enbJHN2PGDDo7O4e1dXZ2MmPGjMl+aTOzhjTunLHAjUBPRBwHkLQFWA4cmczC3q7Pfvaz3HPPPUD5SL6zs5N77rnnvKN8M7OpYtw+ekl3Assi4jPJ408DH4mItal1VgL/FSgBLwF3R8TLybJzwH7gHPCViNha4TVWA6sB5s6d++Gf/exnF/RL3XLLLezYsYOIQBJLlixh+/btF/ScZmaN7EL76FWhbeS7w18D8yLig8DfAE+kls1NXvxPgK9Lev95TxaxKSIKEVFobW3NUNLourq6OHbsGDt37uSNN95g586dHDt2jK6urgt6XjOzi1WWoO8Frk49ngOcSK8QEa9FxNDZzm8CH04tO5H8PA48B3zoAuodV0dHB5s3b2bx4sVMnz6dxYsXs3nzZjo6OibzZc3MGlaWoN8LLJA0X9KlwApg2NUzkt6Teng70J20N0uakdyfBdzMJPftd3d3s2jRomFtixYtoru7ezJf1sysYY0b9BFxDlgLbKcc4E9GxGFJGyUNXUXzOUmHJR0APgesTNrbgGLSvptyH/2kBn1bWxt79uwZ1rZnzx7a2tom82XNzBpWlqtuiIjvA98f0XZ/6v69wL0Vtvs/wL+4wBonZMOGDaxatYrNmzezaNEi9uzZw6pVq9x1Y2ZTVqagv5i0t7cD5aEQuru7aWtro6Oj4812M7OpJndDIJiZTUUXPASCmZldvBz0ZmY556A3M8s5B72ZWc456M3Mcs5Bb2aWcw56M7Occ9CbmeVcboO+9HqJlc+u5NXTr9a7FDOzuspt0Hce7OT5nz9P54HO8Vc2M8uxXAZ96fUSz/Q8QxBs7dnqo3ozm9JyGfSdBzsZjEEABmPQR/VmNqXlLuiHjub7B/sB6B/s91G9mU1pmYJe0jJJRyX1SFpfYflKSSVJ+5PbZ1LL7pJ0LLndVc3iK0kfzQ/xUb2ZTWXjjkcvqQl4FFhCef7YvZK2VZgp6jsRsXbEti3AA0CB8oTi+5Jt+6pSfQUHTh5482h+SP9gP/tP7p+slzQza2hZJh65EehJJvdG0hZgOdnmfr0F2BERp5JtdwDLgK63V+5bWlpa6OvL/n5xiEMIndfe3NzMqVOnLrQcM7OGlSXoZwMvpx73Ah+psN4dkj4GvATcHREvj7Lt7LdZ6zCnPjcAvKsKzzRQhecwM2tcWYL+/MPgcjdM2l8DXRFxVtIa4AngExm3RdJqYDXA3LlzM5QEPPjLbOuZmU1xWU7G9gJXpx7PAU6kV4iI1yLibPLwm8CHs26bbL8pIgoRUWhtbc1au5mZZZAl6PcCCyTNl3QpsALYll5B0ntSD28HupP724GlkpolNQNLkzYzM6uRcbtuIuKcpLWUA7oJeCwiDkvaCBQjYhvwOUm3A+eAU8DKZNtTkh6i/GYBsHHoxKyZmdWGIs7rMq+rQqEQxWKx3mWYmV1UJO2LiEKlZbn7ZqyZmQ3noDczyzkHvZlZzjnozcxyruFOxkoqAT+r0tPNAhpt2ErXlF0j1uWasnFN2VWrrvdGRMUvIjVc0FeTpOJoZ6HrxTVl14h1uaZsXFN2tajLXTdmZjnnoDczy7m8B/2mehdQgWvKrhHrck3ZuKbsJr2uXPfRm5lZ/o/ozcymPAe9mVnO5TLoJT0m6aSkQ/WuJU3SP0p6MZlAvS4jt1XaN5JaJO1IJnDfkQwpXe+aHpT0z6kJ52+rcU1XS9otqVvSYUl/lrTXbV+NUVO999VMST+WdCCp68tJ+3xJ/5Dsq+8kw5zXu6bHJf00ta9uqFVNqdqaJL0g6XvJ40nfT7kMeuBxynPTNqLFEXFDHa/nfZzz9816YGdELAB2Jo/rXRPA15J9dUNEfL/GNZ0DvhARbcBHgT+VdC313Vej1QT13VdngU9ExPXADcAySR8F/ltS1wKgD1jVADUBfDG1r/bXsKYhf8Zbc3ZADfZTLoM+Iv6W8rj4NsIo+2Y55ekfSX5+sgFqqquIeCUink/u/z/Kf5izqeO+GqOmuoqyXycPpye3oDyd6FNJe6331Wg11ZWkOcC/Ab6VPBY12E+5DPoGFsAPJO1L5sltFL8TEa9AOUyAd9e5niFrJR1MunZq2p2UJmke8CHgH2iQfTWiJqjzvkq6I/YDJ4EdwE+AX0TEuWSVXmr8pjSypogY2lcdyb76mqQZtawJ+DrwX4DB5PGV1GA/Oehr6+aI+JfArZQ/dn+s3gU1sG8A76f8sfsV4M/rUYSky4D/BfyniPhVPWoYqUJNdd9XETEQETdQnhf6RqCt0mr1rEnSQuBe4HeBfwW0APfUqh5J/xY4GRH70s0VVq36fnLQ11BEnEh+ngSepvwH0Qh+PjTvb/LzZJ3rISJ+nvyhDlKecL7m+0rSdMqB+j8j4rtJc133VaWaGmFfDYmIXwDPUT6HcIWkoelK5wAn6lzTsqT7KyLiLPDfqe2+uhm4XdI/Alsod9l8nRrsJwd9jUj6LUm/PXSf8kTpjXJV0DbgruT+XcAzdawFOG/C+T+ixvsq6TvdDHRHxFdTi+q2r0arqQH2VaukK5L77wD+kPL5g93Anclqtd5XlWr6v6k3aVHuC6/ZvoqIeyNiTkTMA1YAuyLi31GL/RQRubsBXZQ/wvZT7vNa1QA1vQ84kNwOAxsaZd9Q7ifcCRxLfrY0QE3fBl4EDlIO1/fUuKZFlD9CHwT2J7fb6rmvxqip3vvqg8ALyesfAu5P2t8H/BjoAf4KmNEANe1K9tUh4H8Al9VyX6Xq+zjwvVrtJw+BYGaWc+66MTPLOQe9mVnOOejNzHLOQW9mlnMOejOznHPQm5nlnIPezCzn/j9nXysIb4jTWQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot model performance for comparison\n",
    "_=plt.boxplot(scores, labels=num_features, showmeans=True)\n",
    "_=plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 30 features looks like the right options as it is stable (less std deviation in accuracy of all models).\n",
    "- Also, the accuracy is similar to that of all features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Sort all features based on importance (lower ranking => high importance)\n",
    "# rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=1, verbose=5)\n",
    "# rfe.fit(X_train, y_train)\n",
    "# pd.DataFrame({'features':X.columns.to_list(), 'ranking':rfe.ranking_}).sort_values(by='ranking')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Based on grid search 30 features are being chosen\n",
    "\n",
    "- Selected features\n",
    "          ['dec', 'koi_depth', 'koi_depth_err1', 'koi_depth_err2', 'koi_duration',\n",
    "       'koi_duration_err1', 'koi_duration_err2', 'koi_fpflag_co',\n",
    "       'koi_fpflag_ec', 'koi_fpflag_nt', 'koi_fpflag_ss', 'koi_impact',\n",
    "       'koi_insol', 'koi_insol_err1', 'koi_insol_err2', 'koi_kepmag',\n",
    "       'koi_model_snr', 'koi_period', 'koi_period_err1', 'koi_period_err2',\n",
    "       'koi_prad', 'koi_prad_err1', 'koi_prad_err2', 'koi_steff_err1',\n",
    "       'koi_steff_err2', 'koi_teq', 'koi_time0bk', 'koi_time0bk_err1',\n",
    "       'koi_time0bk_err2', 'ra']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting estimator with 40 features.\n",
      "Fitting estimator with 39 features.\n",
      "Fitting estimator with 38 features.\n",
      "Fitting estimator with 37 features.\n",
      "Fitting estimator with 36 features.\n",
      "Fitting estimator with 35 features.\n",
      "Fitting estimator with 34 features.\n",
      "Fitting estimator with 33 features.\n",
      "Fitting estimator with 32 features.\n",
      "Fitting estimator with 31 features.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RFE(estimator=RandomForestClassifier(), n_features_to_select=30, verbose=5)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=30, verbose=5)\n",
    "rfe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#X_train.columns[rfe.support_]\n",
    "X_train = X_train[X_train.columns[rfe.support_]]\n",
    "X_test = X_test[X_test.columns[rfe.support_]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM without Fine Tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('scale', MinMaxScaler()),\n",
       "                ('model', SVC(class_weight='balanced'))])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create pipeline\n",
    "scaler = MinMaxScaler()\n",
    "#rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=30, verbose=5)\n",
    "model = SVC(class_weight = 'balanced')\n",
    "pipeline = Pipeline(steps=[('scale', scaler), ('model',model)])\n",
    "\n",
    "pipeline.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train accuracy : 0.841502956322716\n",
      "Test accuracy : 0.834096109839817\n"
     ]
    }
   ],
   "source": [
    "print(f\"Train accuracy : {pipeline.score(X_train, y_train)}\")\n",
    "print(f\"Test accuracy : {pipeline.score(X_test, y_test)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GridSearch CV to tune SVM parameters\n",
    "\n",
    "- Mainly 3 parameters to tune in SVM\n",
    "    - kernel\n",
    "    - C (Regularization) Lower C => Simpler bounderines\n",
    "    - Gamma (How far datapoints influences the hyperplane support vectors ?) Higher Gamma => Complicated boundary\n",
    "      The behavior of the model is very sensitive to the gamma parameter. If gamma is too large, the radius of the area of influence of the support vectors only includes the support vector itself and no amount of regularization with C will be able to prevent overfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 24 candidates, totalling 120 fits\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=poly ..............\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=1, model__gamma=scale, model__kernel=poly, score=0.850, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=poly ..............\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.4s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=1, model__gamma=scale, model__kernel=poly, score=0.846, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=poly ..............\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.9s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=1, model__gamma=scale, model__kernel=poly, score=0.848, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=poly ..............\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    1.2s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=1, model__gamma=scale, model__kernel=poly, score=0.860, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=poly ..............\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    1.6s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=1, model__gamma=scale, model__kernel=poly, score=0.861, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=linear, score=0.846, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=linear, score=0.842, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=linear, score=0.845, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=linear, score=0.857, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=linear, score=0.857, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=rbf, score=0.834, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=rbf, score=0.829, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=rbf, score=0.838, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=rbf, score=0.849, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=1, model__gamma=scale, model__kernel=rbf, score=0.846, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=poly, score=0.833, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=poly, score=0.820, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=poly, score=0.840, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=poly, score=0.839, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=poly, score=0.843, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=linear, score=0.846, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=linear, score=0.842, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=linear, score=0.845, total=   0.4s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=linear, score=0.857, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=linear, score=0.857, total=   0.3s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=rbf, score=0.833, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=rbf, score=0.820, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=rbf, score=0.837, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=rbf, score=0.841, total=   0.5s\n",
      "[CV] model__C=1, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=1, model__gamma=0.1, model__kernel=rbf, score=0.841, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=poly ..............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=poly, score=0.867, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=poly ..............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=poly, score=0.865, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=poly ..............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=poly, score=0.862, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=poly ..............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=poly, score=0.885, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=poly ..............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=poly, score=0.878, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=linear, score=0.867, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=linear, score=0.871, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=linear, score=0.867, total=   0.9s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=linear, score=0.885, total=   0.6s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=linear ............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=linear, score=0.871, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=rbf, score=0.855, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=rbf, score=0.847, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=rbf, score=0.855, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=rbf, score=0.869, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=scale, model__kernel=rbf ...............\n",
      "[CV]  model__C=5, model__gamma=scale, model__kernel=rbf, score=0.869, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=poly, score=0.847, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=poly, score=0.843, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=poly, score=0.846, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=poly, score=0.854, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=poly ................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=poly, score=0.854, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=linear, score=0.867, total=   0.3s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=linear, score=0.871, total=   0.3s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=linear, score=0.867, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=linear, score=0.885, total=   0.4s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=linear ..............\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=linear, score=0.871, total=   0.3s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=rbf, score=0.843, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=rbf, score=0.846, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=rbf .................\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=rbf, score=0.846, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=rbf, score=0.859, total=   0.5s\n",
      "[CV] model__C=5, model__gamma=0.1, model__kernel=rbf .................\n",
      "[CV]  model__C=5, model__gamma=0.1, model__kernel=rbf, score=0.857, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=poly .............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=poly, score=0.869, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=poly .............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=poly, score=0.872, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=poly .............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=poly, score=0.870, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=poly .............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=poly, score=0.887, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=poly .............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=poly, score=0.883, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=linear ...........\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=linear, score=0.871, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=linear ...........\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=linear, score=0.879, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=linear ...........\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=linear, score=0.873, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=linear ...........\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=linear, score=0.889, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=linear ...........\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=linear, score=0.877, total=   0.3s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=rbf ..............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=rbf, score=0.862, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=rbf ..............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=rbf, score=0.857, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=rbf ..............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=rbf, score=0.860, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=rbf ..............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=rbf, score=0.878, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=scale, model__kernel=rbf ..............\n",
      "[CV]  model__C=10, model__gamma=scale, model__kernel=rbf, score=0.872, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=poly ...............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=poly, score=0.850, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=poly ...............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=poly, score=0.848, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=poly ...............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=poly, score=0.858, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=poly ...............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=poly, score=0.866, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=poly ...............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=poly, score=0.863, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=linear .............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=linear, score=0.871, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=linear .............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=linear, score=0.879, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=linear .............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=linear, score=0.873, total=   0.4s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=linear .............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=linear, score=0.889, total=   0.6s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=linear .............\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=linear, score=0.877, total=   0.8s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=rbf ................\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=rbf, score=0.853, total=   0.7s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=rbf ................\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=rbf, score=0.850, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=rbf ................\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=rbf, score=0.859, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=rbf ................\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=rbf, score=0.872, total=   0.5s\n",
      "[CV] model__C=10, model__gamma=0.1, model__kernel=rbf ................\n",
      "[CV]  model__C=10, model__gamma=0.1, model__kernel=rbf, score=0.868, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=poly ............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=poly, score=0.878, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=poly ............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=poly, score=0.892, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=poly ............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=poly, score=0.879, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=poly ............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=poly, score=0.884, total=   0.7s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=poly ............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=poly, score=0.887, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=linear ..........\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=linear, score=0.881, total=   0.6s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=linear ..........\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=linear, score=0.893, total=   0.7s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=linear ..........\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=linear, score=0.880, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=linear ..........\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=linear, score=0.895, total=   0.8s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=linear ..........\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=linear, score=0.888, total=   0.6s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=rbf .............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=rbf, score=0.874, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=rbf .............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=rbf, score=0.890, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=rbf .............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=rbf, score=0.880, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=rbf .............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=rbf, score=0.885, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=scale, model__kernel=rbf .............\n",
      "[CV]  model__C=100, model__gamma=scale, model__kernel=rbf, score=0.887, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=poly ..............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=poly, score=0.868, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=poly ..............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=poly, score=0.875, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=poly ..............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=poly, score=0.880, total=   0.3s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=poly ..............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=poly, score=0.890, total=   0.3s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=poly ..............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=poly, score=0.890, total=   0.3s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=linear ............\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=linear, score=0.881, total=   0.6s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=linear ............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=linear, score=0.893, total=   0.8s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=linear ............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=linear, score=0.880, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=linear ............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=linear, score=0.895, total=   0.8s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=linear ............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=linear, score=0.888, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=rbf ...............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=rbf, score=0.868, total=   0.5s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=rbf ...............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=rbf, score=0.875, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=rbf ...............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=rbf, score=0.882, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=rbf ...............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=rbf, score=0.889, total=   0.4s\n",
      "[CV] model__C=100, model__gamma=0.1, model__kernel=rbf ...............\n",
      "[CV]  model__C=100, model__gamma=0.1, model__kernel=rbf, score=0.886, total=   0.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done 120 out of 120 | elapsed:   54.6s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[('scale', MinMaxScaler()),\n",
       "                                       ('model',\n",
       "                                        SVC(class_weight='balanced'))]),\n",
       "             param_grid={'model__C': [1, 5, 10, 100],\n",
       "                         'model__gamma': ['scale', 0.1],\n",
       "                         'model__kernel': ['poly', 'linear', 'rbf']},\n",
       "             scoring='accuracy', verbose=5)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "param_grid = {'model__C': [1, 5, 10, 100],\n",
    "              'model__gamma': ['scale', 0.1], #if gamma='scale' (default) is passed then it uses 1 / (n_features * X.var()) as value of gamma, \n",
    "                                                                #if ‘auto’, uses 1 / n_features.\n",
    "              'model__kernel': ['poly','linear', 'rbf']\n",
    "               }\n",
    "\n",
    "# create pipeline\n",
    "scaler = MinMaxScaler()\n",
    "#rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=30, verbose=5)\n",
    "model = SVC(class_weight = 'balanced') #Multipliers of parameter C for each class. Computed based on the class_weight parameter.\n",
    "pipeline = Pipeline(steps=[('scale', scaler), ('model',model)])\n",
    "\n",
    "gcv = GridSearchCV(\n",
    "        pipeline, \n",
    "        param_grid,\n",
    "        cv=5,\n",
    "        scoring = 'accuracy',\n",
    "        verbose=5\n",
    "    )\n",
    "\n",
    "gcv.fit(X_train, y_train) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model__C': 100, 'model__gamma': 'scale', 'model__kernel': 'linear'}"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "0.8874706190555891"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#pipeline.get_params().keys()\n",
    "gcv.best_params_\n",
    "gcv.best_score_\n",
    "#{'model__C': 100, 'model__gamma': 'scale', 'model__kernel': 'linear'}\n",
    "#0.8874706190555891"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the final model on test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MinMaxScaler()"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "SVC(C=100, class_weight='balanced', kernel='linear')"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.883\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7fa2216c5ee0>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Confusion Matrix (Test Data) of SVM')"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# rfe = RFE(estimator=RandomForestClassifier(n_estimators=100), n_features_to_select=30, verbose=5)\n",
    "# rfe.fit(X_train, y_train)\n",
    "# X_train_selected_features = rfe.transform(X_train)\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train_scaled = scaler.transform(X_train)\n",
    "\n",
    "model = SVC(class_weight = 'balanced', C = 100, gamma = 'scale', kernel = 'linear') #gamma is not applicable to linear kernel\n",
    "model.fit(X_train_scaled, y_train)\n",
    "\n",
    "#On test\n",
    "X_test_scaled = scaler.transform(X_test)\n",
    "y_pred = model.predict(X_test_scaled)\n",
    "\n",
    "#Final test accuracy\n",
    "print('Accuracy: %.3f' % model.score(X_test_scaled, y_test))\n",
    "\n",
    "fig, ax = plt.subplots(1,1)\n",
    "plot_confusion_matrix(model, \n",
    "             X_test_scaled,\n",
    "             y_test,\n",
    "             values_format='d',\n",
    "             ax=ax)\n",
    "\n",
    "ax.set_title(\"Confusion Matrix (Test Data) of SVM\", fontsize=16)\n",
    "plt.savefig(\"../Images/svm_cn.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                precision    recall  f1-score   support\n",
      "\n",
      "     CANDIDATE       0.81      0.69      0.74       414\n",
      "     CONFIRMED       0.76      0.84      0.79       461\n",
      "FALSE POSITIVE       0.99      1.00      0.99       873\n",
      "\n",
      "      accuracy                           0.88      1748\n",
      "     macro avg       0.85      0.84      0.84      1748\n",
      "  weighted avg       0.88      0.88      0.88      1748\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Bincy_Narath_SVM.sav']"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import joblib\n",
    "filename = 'Bincy_Narath_SVM.sav'\n",
    "joblib.dump(model, filename)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.1 64-bit ('base': conda)",
   "language": "python",
   "name": "python38164bitbasecondaa2e9191beccf444484f1e2cc79ad9f61"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}