FYP_final code_V1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.metrics import *\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.pipeline import Pipeline \n",
    "from sklearn.compose import ColumnTransformer\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy import stats\n",
    "from sklearn.model_selection import LeaveOneOut\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.utils.data import DataLoader, TensorDataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Size</th>\n",
       "      <th>Dose</th>\n",
       "      <th>P_density</th>\n",
       "      <th>CSKKKK</th>\n",
       "      <th>QHREDGS</th>\n",
       "      <th>RKDVY</th>\n",
       "      <th>WKYMVM</th>\n",
       "      <th>CLPFFD</th>\n",
       "      <th>QEINSSY</th>\n",
       "      <th>fold_change</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.723886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.699132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.749064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.762759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.690541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.280688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.002028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.386585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.443380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.123173</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>126 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Size  Dose  P_density  CSKKKK  QHREDGS  RKDVY  WKYMVM  CLPFFD  QEINSSY  \\\n",
       "0    HNP  1000       1.71       2        0      0       0       0        0   \n",
       "1    HNP  1000       1.71       0        2      0       0       0        0   \n",
       "2    HNP  1000       1.71       0        0      2       0       0        0   \n",
       "3    HNP  1000       1.71       0        0      0       2       0        0   \n",
       "4    HNP  1000       1.71       0        0      0       0       2        0   \n",
       "..   ...   ...        ...     ...      ...    ...     ...     ...      ...   \n",
       "121  HNP   100      19.48       0        0      1       0       1        0   \n",
       "122  HNP   100      19.48       0        0      1       0       0        1   \n",
       "123  HNP   100      19.48       0        0      0       1       1        0   \n",
       "124  HNP   100      19.48       0        0      0       1       0        1   \n",
       "125  HNP   100      19.48       0        0      0       0       1        1   \n",
       "\n",
       "     fold_change  \n",
       "0      -0.723886  \n",
       "1      -0.699132  \n",
       "2      -0.749064  \n",
       "3      -0.762759  \n",
       "4      -0.690541  \n",
       "..           ...  \n",
       "121    -0.280688  \n",
       "122     0.002028  \n",
       "123    -0.386585  \n",
       "124    -0.443380  \n",
       "125    -0.123173  \n",
       "\n",
       "[126 rows x 10 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_in = pd.read_csv('finaldatabase_HNP.csv')\n",
    "df_in\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 2.,  6.,  4., 11.,  8.,  5.,  1.,  1.,  0.,  2.,  6.,  1.,  0.,\n",
       "         2.,  2.,  5.,  2.,  6.,  2.,  2.,  2.,  3.,  4.,  6.,  1.,  2.,\n",
       "         1.,  4.,  3.,  1.,  5.,  2.,  1.,  1.,  1.,  2.,  2.,  2.,  2.,\n",
       "         1.,  1.,  0.,  0.,  1.,  0.,  2.,  1.,  0.,  0.,  0.,  0.,  0.,\n",
       "         0.,  0.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  1.,\n",
       "         0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  1.,\n",
       "         0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  1.,\n",
       "         0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  1.]),\n",
       " array([-0.78266417, -0.76113787, -0.73961157, -0.71808528, -0.69655898,\n",
       "        -0.67503268, -0.65350638, -0.63198008, -0.61045378, -0.58892749,\n",
       "        -0.56740119, -0.54587489, -0.52434859, -0.50282229, -0.481296  ,\n",
       "        -0.4597697 , -0.4382434 , -0.4167171 , -0.3951908 , -0.3736645 ,\n",
       "        -0.35213821, -0.33061191, -0.30908561, -0.28755931, -0.26603301,\n",
       "        -0.24450671, -0.22298042, -0.20145412, -0.17992782, -0.15840152,\n",
       "        -0.13687522, -0.11534892, -0.09382263, -0.07229633, -0.05077003,\n",
       "        -0.02924373, -0.00771743,  0.01380887,  0.03533516,  0.05686146,\n",
       "         0.07838776,  0.09991406,  0.12144036,  0.14296666,  0.16449295,\n",
       "         0.18601925,  0.20754555,  0.22907185,  0.25059815,  0.27212444,\n",
       "         0.29365074,  0.31517704,  0.33670334,  0.35822964,  0.37975594,\n",
       "         0.40128223,  0.42280853,  0.44433483,  0.46586113,  0.48738743,\n",
       "         0.50891373,  0.53044002,  0.55196632,  0.57349262,  0.59501892,\n",
       "         0.61654522,  0.63807152,  0.65959781,  0.68112411,  0.70265041,\n",
       "         0.72417671,  0.74570301,  0.76722931,  0.7887556 ,  0.8102819 ,\n",
       "         0.8318082 ,  0.8533345 ,  0.8748608 ,  0.89638709,  0.91791339,\n",
       "         0.93943969,  0.96096599,  0.98249229,  1.00401859,  1.02554488,\n",
       "         1.04707118,  1.06859748,  1.09012378,  1.11165008,  1.13317638,\n",
       "         1.15470267,  1.17622897,  1.19775527,  1.21928157,  1.24080787,\n",
       "         1.26233417,  1.28386046,  1.30538676,  1.32691306,  1.34843936,\n",
       "         1.36996566]),\n",
       " <BarContainer object of 100 artists>)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh8AAAGdCAYAAACyzRGfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAXqElEQVR4nO3dfWxV9f3A8U8BKehoEZWHRhR8mCgg4gMEXXRGMubQyD+bOmYYc+IyNmUsaLtMCXNaNcaQOaLORHGJj0umLjIxhumIE5EnN0WnoqidrrDN2QJudaPn98cvayxPAt77aS+8Xsn54577vef7PT294e3pra0qiqIIAIAkPbp6AQDA/kV8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpenX1ArbV3t4e77//fvTr1y+qqqq6ejkAwG4oiiI2bdoUdXV10aPHru9tdLv4eP/992Po0KFdvQwAYC80NTXF4Ycfvssx3S4++vXrFxH/v/iampouXg0AsDtaW1tj6NChHf+O70q3i4///ailpqZGfABAhdmdj0z4wCkAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpenX1AirBsPpF2+17+8bJXbASAKh87nwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKn2OD6WLl0a559/ftTV1UVVVVU8+uijnZ4viiKuvfbaGDJkSPTt2zcmTpwYb7zxRqnWCwBUuD2Ojy1btsSYMWNiwYIFO3z+5ptvjp/97Gdxxx13xPLly+Oggw6KSZMmxb///e/PvFgAoPL12tMXnHvuuXHuuefu8LmiKGL+/Pnx4x//OC644IKIiPjlL38ZgwYNikcffTQuuuiiz7ZaAKDilfQzH+vXr4/m5uaYOHFix77a2toYP358LFu2bIevaWtri9bW1k4bALDvKml8NDc3R0TEoEGDOu0fNGhQx3PbamxsjNra2o5t6NChpVwSANDNdPlvuzQ0NERLS0vH1tTU1NVLAgDKqKTxMXjw4IiI2LBhQ6f9GzZs6HhuW9XV1VFTU9NpAwD2XSWNj+HDh8fgwYNjyZIlHftaW1tj+fLlMWHChFJOBQBUqD3+bZfNmzfHunXrOh6vX78+XnzxxRgwYEAcccQRMWvWrPjpT38axx57bAwfPjyuueaaqKuriylTppRy3QBAhdrj+Fi5cmWcffbZHY9nz54dERHTpk2LhQsXxlVXXRVbtmyJGTNmxIcffhhf+MIXYvHixdGnT5/SrRoAqFhVRVEUXb2IT2ptbY3a2tpoaWnpNp//GFa/aLt9b984uQtWAgDd0578+93lv+0CAOxfxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpxAcAkEp8AACpSh4fW7dujWuuuSaGDx8effv2jaOPPjquu+66KIqi1FMBABWoV6kPeNNNN8Xtt98e9957b4wcOTJWrlwZ06dPj9ra2rjiiitKPR0AUGFKHh/PPfdcXHDBBTF58uSIiBg2bFg88MAD8cILL5R6KgCgApX8xy6nn356LFmyJF5//fWIiPjjH/8Yzz77bJx77rk7HN/W1hatra2dNgBg31XyOx/19fXR2toaI0aMiJ49e8bWrVvj+uuvj6lTp+5wfGNjY8ybN6/Uyyi7YfWLOj1++8bJXbQSAKgsJb/z8fDDD8d9990X999/f6xevTruvffeuOWWW+Lee+/d4fiGhoZoaWnp2Jqamkq9JACgGyn5nY85c+ZEfX19XHTRRRERMXr06HjnnXeisbExpk2btt346urqqK6uLvUyAIBuquR3Pj766KPo0aPzYXv27Bnt7e2lngoAqEAlv/Nx/vnnx/XXXx9HHHFEjBw5MtasWRO33nprfOtb3yr1VABABSp5fNx2221xzTXXxHe/+93YuHFj1NXVxeWXXx7XXnttqacCACpQyeOjX79+MX/+/Jg/f36pDw0A7AP8bRcAIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIJX4AABSiQ8AIFVZ4uO9996Lb3zjG3HIIYdE3759Y/To0bFy5cpyTAUAVJhepT7gP//5zzjjjDPi7LPPjieeeCIOO+yweOONN+Lggw8u9VQAQAUqeXzcdNNNMXTo0Ljnnns69g0fPrzU0wAAFarkP3b5zW9+E6eeemp89atfjYEDB8bYsWPjrrvu2un4tra2aG1t7bQBAPuukt/5eOutt+L222+P2bNnx49+9KNYsWJFXHHFFdG7d++YNm3aduMbGxtj3rx5pV7GbhtWv2i7fW/fOLkLVrJz264xe32Z83f1uQJQfiW/89He3h4nn3xy3HDDDTF27NiYMWNGXHbZZXHHHXfscHxDQ0O0tLR0bE1NTaVeEgDQjZQ8PoYMGRInnHBCp33HH398vPvuuzscX11dHTU1NZ02AGDfVfL4OOOMM+K1117rtO/111+PI488stRTAQAVqOTx8YMf/CCef/75uOGGG2LdunVx//33xy9+8YuYOXNmqacCACpQyePjtNNOi0ceeSQeeOCBGDVqVFx33XUxf/78mDp1aqmnAgAqUMl/2yUi4rzzzovzzjuvHIcGACqcv+0CAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKTq1dUL2FcMq1+03b63b5zcBSvZM9uuuxLWXC6Veg0BKo07HwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAqrLHx4033hhVVVUxa9asck8FAFSAssbHihUr4s4774wTTzyxnNMAABWkbPGxefPmmDp1atx1111x8MEHl2saAKDClC0+Zs6cGZMnT46JEyfuclxbW1u0trZ22gCAfVevchz0wQcfjNWrV8eKFSs+dWxjY2PMmzevHMvYa8PqF3X1EvYL5fw6b3vst2+cXLa5ANgzJb/z0dTUFFdeeWXcd9990adPn08d39DQEC0tLR1bU1NTqZcEAHQjJb/zsWrVqti4cWOcfPLJHfu2bt0aS5cujZ///OfR1tYWPXv27Hiuuro6qqurS70MAKCbKnl8nHPOOfHSSy912jd9+vQYMWJEXH311Z3CAwDY/5Q8Pvr16xejRo3qtO+ggw6KQw45ZLv9AMD+x//hFABIVZbfdtnWM888kzENAFAB3PkAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFKJDwAglfgAAFL16uoFsP8YVr+oW8399o2Tu2AlALjzAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkEh8AQCrxAQCkKnl8NDY2xmmnnRb9+vWLgQMHxpQpU+K1114r9TQAQIUqeXz8/ve/j5kzZ8bzzz8fTz31VPznP/+JL33pS7Fly5ZSTwUAVKBepT7g4sWLOz1euHBhDBw4MFatWhVnnnlmqacDACpMyeNjWy0tLRERMWDAgB0+39bWFm1tbR2PW1tby70kAKALlTU+2tvbY9asWXHGGWfEqFGjdjimsbEx5s2bV85ldDKsflHaXHujlOvbm2Ptzmu6+mu47fxv3zi5289Vqq9ZOc91W5lfZ2D/Utbfdpk5c2a8/PLL8eCDD+50TENDQ7S0tHRsTU1N5VwSANDFynbn43vf+148/vjjsXTp0jj88MN3Oq66ujqqq6vLtQwAoJspeXwURRHf//7345FHHolnnnkmhg8fXuopAIAKVvL4mDlzZtx///3x2GOPRb9+/aK5uTkiImpra6Nv376lng4AqDAl/8zH7bffHi0tLfHFL34xhgwZ0rE99NBDpZ4KAKhAZfmxCwDAzvjbLgBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAKvEBAKQSHwBAqqqiKIquXsQntba2Rm1tbbS0tERNTU3Jjz+sflHJj/lZvH3j5E6P93Z92x7nsxyLyraj74Vtbfu9sbffP7szV1fa0TmU6uuzN3Pt7Xoq0f58rt3tPLOuxZ78++3OBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQSnwAAKnEBwCQqmzxsWDBghg2bFj06dMnxo8fHy+88EK5pgIAKkhZ4uOhhx6K2bNnx9y5c2P16tUxZsyYmDRpUmzcuLEc0wEAFaQs8XHrrbfGZZddFtOnT48TTjgh7rjjjjjwwAPj7rvvLsd0AEAF6VXqA3788cexatWqaGho6NjXo0ePmDhxYixbtmy78W1tbdHW1tbxuKWlJSIiWltbS720iIhob/uoLMfdW9ue596ub0dfr+52ruTYnffOtt8be/v9U673aans6BxK9fXZm7n2dj2VaH8+1+52nlnX4n/HLIri0wcXJfbee+8VEVE899xznfbPmTOnGDdu3Hbj586dW0SEzWaz2Wy2fWBramr61FYo+Z2PPdXQ0BCzZ8/ueNze3h4ffPBBHHLIIVFVVdWFK6tsra2tMXTo0GhqaoqampquXg474Bp1f65R9+b6dC9FUcSmTZuirq7uU8eWPD4OPfTQ6NmzZ2zYsKHT/g0bNsTgwYO3G19dXR3V1dWd9vXv37/Uy9pv1dTUeFN2c65R9+cadW+uT/dRW1u7W+NK/oHT3r17xymnnBJLlizp2Nfe3h5LliyJCRMmlHo6AKDClOXHLrNnz45p06bFqaeeGuPGjYv58+fHli1bYvr06eWYDgCoIGWJjwsvvDD+9re/xbXXXhvNzc1x0kknxeLFi2PQoEHlmI4dqK6ujrlz5273Iy26D9eo+3ONujfXp3JVFcXu/E4MAEBp+NsuAEAq8QEApBIfAEAq8QEApBIf+5APPvggpk6dGjU1NdG/f/+49NJLY/Pmzbt8zRe/+MWoqqrqtH3nO99JWvG+b8GCBTFs2LDo06dPjB8/Pl544YVdjv/Vr34VI0aMiD59+sTo0aPjt7/9bdJK9197co0WLly43fulT58+iavdvyxdujTOP//8qKuri6qqqnj00Uc/9TXPPPNMnHzyyVFdXR3HHHNMLFy4sOzrZM+Jj33I1KlTY+3atfHUU0/F448/HkuXLo0ZM2Z86usuu+yy+Otf/9qx3XzzzQmr3fc99NBDMXv27Jg7d26sXr06xowZE5MmTYqNGzfucPxzzz0XF198cVx66aWxZs2amDJlSkyZMiVefvnl5JXvP/b0GkX8//9N85Pvl3feeSdxxfuXLVu2xJgxY2LBggW7NX79+vUxefLkOPvss+PFF1+MWbNmxbe//e148skny7xS9lhp/pwcXe2VV14pIqJYsWJFx74nnniiqKqqKt57772dvu6ss84qrrzyyoQV7n/GjRtXzJw5s+Px1q1bi7q6uqKxsXGH47/2ta8VkydP7rRv/PjxxeWXX17Wde7P9vQa3XPPPUVtbW3S6vikiCgeeeSRXY656qqripEjR3bad+GFFxaTJk0q48rYG+587COWLVsW/fv3j1NPPbVj38SJE6NHjx6xfPnyXb72vvvui0MPPTRGjRoVDQ0N8dFHn/6n1Nm1jz/+OFatWhUTJ07s2NejR4+YOHFiLFu2bIevWbZsWafxERGTJk3a6Xg+m725RhERmzdvjiOPPDKGDh0aF1xwQaxduzZjuewG76HK0eV/1ZbSaG5ujoEDB3ba16tXrxgwYEA0Nzfv9HVf//rX48gjj4y6urr405/+FFdffXW89tpr8etf/7rcS96n/f3vf4+tW7du93/1HTRoUPz5z3/e4Wuam5t3OH5X14+9tzfX6Ljjjou77747TjzxxGhpaYlbbrklTj/99Fi7dm0cfvjhGctmF3b2HmptbY1//etf0bdv3y5aGdsSH91cfX193HTTTbsc8+qrr+718T/5mZDRo0fHkCFD4pxzzok333wzjj766L0+LuyLJkyY0OkPZJ5++ulx/PHHx5133hnXXXddF64MKov46OZ++MMfxje/+c1djjnqqKNi8ODB231I7r///W988MEHMXjw4N2eb/z48RERsW7dOvHxGRx66KHRs2fP2LBhQ6f9GzZs2On1GDx48B6N57PZm2u0rQMOOCDGjh0b69atK8cS2UM7ew/V1NS469HN+MxHN3fYYYfFiBEjdrn17t07JkyYEB9++GGsWrWq47W/+93vor29vSModseLL74YERFDhgwp9ansV3r37h2nnHJKLFmypGNfe3t7LFmypNN/OX/ShAkTOo2PiHjqqad2Op7PZm+u0ba2bt0aL730kvdLN+E9VEG6+hOvlM6Xv/zlYuzYscXy5cuLZ599tjj22GOLiy++uOP5v/zlL8Vxxx1XLF++vCiKoli3bl3xk5/8pFi5cmWxfv364rHHHiuOOuqo4swzz+yqU9inPPjgg0V1dXWxcOHC4pVXXilmzJhR9O/fv2hubi6KoiguueSSor6+vmP8H/7wh6JXr17FLbfcUrz66qvF3LlziwMOOKB46aWXuuoU9nl7eo3mzZtXPPnkk8Wbb75ZrFq1qrjooouKPn36FGvXru2qU9inbdq0qVizZk2xZs2aIiKKW2+9tVizZk3xzjvvFEVRFPX19cUll1zSMf6tt94qDjzwwGLOnDnFq6++WixYsKDo2bNnsXjx4q46BXZCfOxD/vGPfxQXX3xx8bnPfa6oqakppk+fXmzatKnj+fXr1xcRUTz99NNFURTFu+++W5x55pnFgAEDiurq6uKYY44p5syZU7S0tHTRGex7brvttuKII44oevfuXYwbN654/vnnO54766yzimnTpnUa//DDDxef//zni969excjR44sFi1alLzi/c+eXKNZs2Z1jB00aFDxla98pVi9enUXrHr/8PTTTxcRsd32v2sybdq04qyzztruNSeddFLRu3fv4qijjiruueee9HXz6aqKoii68MYLALCf8ZkPACCV+AAAUokPACCV+AAAUokPACCV+AAAUokPACCV+AAAUokPACCV+AAAUokPACCV+AAAUv0fi7773SXpVegAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(df_in.fold_change,  bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Size</th>\n",
       "      <th>Dose</th>\n",
       "      <th>P_density</th>\n",
       "      <th>CSKKKK</th>\n",
       "      <th>QHREDGS</th>\n",
       "      <th>RKDVY</th>\n",
       "      <th>WKYMVM</th>\n",
       "      <th>CLPFFD</th>\n",
       "      <th>QEINSSY</th>\n",
       "      <th>fold_change</th>\n",
       "      <th>Polarization_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.723886</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.699132</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.749064</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.762759</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.690541</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.280688</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.002028</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.386585</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.443380</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.123173</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>126 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Size  Dose  P_density  CSKKKK  QHREDGS  RKDVY  WKYMVM  CLPFFD  QEINSSY  \\\n",
       "0    HNP  1000       1.71       2        0      0       0       0        0   \n",
       "1    HNP  1000       1.71       0        2      0       0       0        0   \n",
       "2    HNP  1000       1.71       0        0      2       0       0        0   \n",
       "3    HNP  1000       1.71       0        0      0       2       0        0   \n",
       "4    HNP  1000       1.71       0        0      0       0       2        0   \n",
       "..   ...   ...        ...     ...      ...    ...     ...     ...      ...   \n",
       "121  HNP   100      19.48       0        0      1       0       1        0   \n",
       "122  HNP   100      19.48       0        0      1       0       0        1   \n",
       "123  HNP   100      19.48       0        0      0       1       1        0   \n",
       "124  HNP   100      19.48       0        0      0       1       0        1   \n",
       "125  HNP   100      19.48       0        0      0       0       1        1   \n",
       "\n",
       "     fold_change  Polarization_group  \n",
       "0      -0.723886                   0  \n",
       "1      -0.699132                   0  \n",
       "2      -0.749064                   0  \n",
       "3      -0.762759                   0  \n",
       "4      -0.690541                   0  \n",
       "..           ...                 ...  \n",
       "121    -0.280688                   0  \n",
       "122     0.002028                   0  \n",
       "123    -0.386585                   0  \n",
       "124    -0.443380                   0  \n",
       "125    -0.123173                   0  \n",
       "\n",
       "[126 rows x 11 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_in = df_in[['Size','Dose','P_density','CSKKKK','QHREDGS','RKDVY','WKYMVM','CLPFFD','QEINSSY','fold_change']]\n",
    "df_in['Polarization_group'] = 0\n",
    "df_in"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "for i in range(len(df_in)):\n",
    "  if df_in.fold_change[i]<0:\n",
    "    df_in.Polarization_group[i] = 1\n",
    "  elif df_in.fold_change[i]>0:\n",
    "    df_in.Polarization_group[i] = 0"
   ]
  },
  {
   "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>Size</th>\n",
       "      <th>Dose</th>\n",
       "      <th>P_density</th>\n",
       "      <th>CSKKKK</th>\n",
       "      <th>QHREDGS</th>\n",
       "      <th>RKDVY</th>\n",
       "      <th>WKYMVM</th>\n",
       "      <th>CLPFFD</th>\n",
       "      <th>QEINSSY</th>\n",
       "      <th>Polarization_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HNP</td>\n",
       "      <td>1000</td>\n",
       "      <td>1.71</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>HNP</td>\n",
       "      <td>100</td>\n",
       "      <td>19.48</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>126 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Size  Dose  P_density  CSKKKK  QHREDGS  RKDVY  WKYMVM  CLPFFD  QEINSSY  \\\n",
       "0    HNP  1000       1.71       2        0      0       0       0        0   \n",
       "1    HNP  1000       1.71       0        2      0       0       0        0   \n",
       "2    HNP  1000       1.71       0        0      2       0       0        0   \n",
       "3    HNP  1000       1.71       0        0      0       2       0        0   \n",
       "4    HNP  1000       1.71       0        0      0       0       2        0   \n",
       "..   ...   ...        ...     ...      ...    ...     ...     ...      ...   \n",
       "121  HNP   100      19.48       0        0      1       0       1        0   \n",
       "122  HNP   100      19.48       0        0      1       0       0        1   \n",
       "123  HNP   100      19.48       0        0      0       1       1        0   \n",
       "124  HNP   100      19.48       0        0      0       1       0        1   \n",
       "125  HNP   100      19.48       0        0      0       0       1        1   \n",
       "\n",
       "     Polarization_group  \n",
       "0                     1  \n",
       "1                     1  \n",
       "2                     1  \n",
       "3                     1  \n",
       "4                     1  \n",
       "..                  ...  \n",
       "121                   1  \n",
       "122                   0  \n",
       "123                   1  \n",
       "124                   1  \n",
       "125                   1  \n",
       "\n",
       "[126 rows x 10 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df_in[['Size','Dose','P_density','CSKKKK','QHREDGS','RKDVY','WKYMVM','CLPFFD','QEINSSY','Polarization_group']]\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(104, 9)\n",
      "(22, 9)\n",
      "(104, 1)\n",
      "(22, 1)\n"
     ]
    },
    {
     "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>Polarization_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Polarization_group\n",
       "16                    1\n",
       "90                    1\n",
       "92                    1\n",
       "6                     1\n",
       "121                   1\n",
       "..                  ...\n",
       "19                    1\n",
       "14                    1\n",
       "50                    0\n",
       "99                    1\n",
       "7                     1\n",
       "\n",
       "[104 rows x 1 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = pd.DataFrame(df, columns=['Polarization_group']) \n",
    "X = df.drop(columns=['Polarization_group'])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
    "                                                    test_size=0.1667, \n",
    "                                                    random_state=0, #We put random state to keep the same selection\n",
    "                                                    stratify=y) \n",
    "                             \n",
    "print(X_train.shape)\n",
    "print(X_test.shape)\n",
    "print(y_train.shape)\n",
    "print(y_test.shape)\n",
    "\n",
    "X_train\n",
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocessing\n",
    "\n",
    "## Identification of numerical and categorical columns\n",
    "numeric_cols = X_train.select_dtypes(exclude=['object', 'category']).columns.to_list()\n",
    "cat_cols = X_train.select_dtypes(include=['object', 'category']).columns.to_list()\n",
    "\n",
    "## Transformations for numerical variables\n",
    "numeric_transformer = Pipeline(steps=[('scaler', StandardScaler())])\n",
    "\n",
    "## Transformations for categorical variables\n",
    "categorical_transformer = Pipeline(steps=[('onehot', OneHotEncoder())])\n",
    "\n",
    "## Preprocessor\n",
    "ct = ColumnTransformer(\n",
    "                    transformers=[\n",
    "                        ('numeric', numeric_transformer, numeric_cols),\n",
    "                        ('cat', categorical_transformer, cat_cols)\n",
    "                    ],\n",
    "                    remainder='passthrough'\n",
    "                )\n",
    " \n",
    "ct.fit(X_train) # Fit all transformers using X training data. \n",
    "#We put the content inside the scaler.\n",
    "\n",
    "#Obtain a list with the names of the new columns after doing the one-hot encoding.\n",
    "colnames = pd.get_dummies(X_train).columns.to_list()\n",
    "\n",
    "scaledTrain_df = pd.DataFrame(ct.transform(X_train), # Transform X separately by \n",
    "                              #each transformer, concatenate results.\n",
    "                              index=X_train.index,\n",
    "                              columns=colnames) \n",
    "\n",
    "scaledValidation_df = pd.DataFrame(ct.transform(X_test), # Transform X separately \n",
    "                                   #by each transformer, concatenate results.\n",
    "                                   index=X_test.index,\n",
    "                                   columns=colnames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(104, 9)\n",
      "(22, 9)\n",
      "(104, 1)\n",
      "(22, 1)\n"
     ]
    },
    {
     "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>Dose</th>\n",
       "      <th>P_density</th>\n",
       "      <th>CSKKKK</th>\n",
       "      <th>QHREDGS</th>\n",
       "      <th>RKDVY</th>\n",
       "      <th>WKYMVM</th>\n",
       "      <th>CLPFFD</th>\n",
       "      <th>QEINSSY</th>\n",
       "      <th>Size_HNP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>-1.161719</td>\n",
       "      <td>-0.634083</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>1.245086</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>1.299137</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>1.417620</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>1.299137</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>-1.161719</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>1.417620</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>-0.96225</td>\n",
       "      <td>1.299137</td>\n",
       "      <td>-0.634083</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>1.245086</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>-1.161719</td>\n",
       "      <td>-0.634083</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>1.196975</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>-1.161719</td>\n",
       "      <td>-0.634083</td>\n",
       "      <td>1.417620</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>1.196975</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>0.040320</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>1.299137</td>\n",
       "      <td>-0.634083</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.03923</td>\n",
       "      <td>-1.161719</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>-0.548146</td>\n",
       "      <td>1.097476</td>\n",
       "      <td>-0.606069</td>\n",
       "      <td>-0.553372</td>\n",
       "      <td>-0.581388</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Dose  P_density    CSKKKK   QHREDGS     RKDVY    WKYMVM    CLPFFD  \\\n",
       "16   1.03923  -1.161719 -0.634083 -0.548146  1.097476 -0.606069  1.245086   \n",
       "90   1.03923   1.299137  1.056806  1.417620 -0.606069 -0.606069 -0.553372   \n",
       "92   1.03923   1.299137  1.056806 -0.548146 -0.606069  1.097476 -0.553372   \n",
       "6    1.03923  -1.161719  1.056806  1.417620 -0.606069 -0.606069 -0.553372   \n",
       "121 -0.96225   1.299137 -0.634083 -0.548146  1.097476 -0.606069  1.245086   \n",
       "..       ...        ...       ...       ...       ...       ...       ...   \n",
       "19   1.03923  -1.161719 -0.634083 -0.548146 -0.606069  1.097476 -0.553372   \n",
       "14   1.03923  -1.161719 -0.634083  1.417620 -0.606069 -0.606069 -0.553372   \n",
       "50   1.03923   0.040320  1.056806 -0.548146 -0.606069  1.097476 -0.553372   \n",
       "99   1.03923   1.299137 -0.634083 -0.548146  1.097476  1.097476 -0.553372   \n",
       "7    1.03923  -1.161719  1.056806 -0.548146  1.097476 -0.606069 -0.553372   \n",
       "\n",
       "      QEINSSY  Size_HNP  \n",
       "16  -0.581388       1.0  \n",
       "90  -0.581388       1.0  \n",
       "92  -0.581388       1.0  \n",
       "6   -0.581388       1.0  \n",
       "121 -0.581388       1.0  \n",
       "..        ...       ...  \n",
       "19   1.196975       1.0  \n",
       "14   1.196975       1.0  \n",
       "50  -0.581388       1.0  \n",
       "99  -0.581388       1.0  \n",
       "7   -0.581388       1.0  \n",
       "\n",
       "[104 rows x 9 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(scaledTrain_df.shape)\n",
    "print(scaledValidation_df.shape)\n",
    "\n",
    "print(y_train.shape)\n",
    "print(y_test.shape)\n",
    "\n",
    "scaledTrain_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean accuracy in the training: 0.8557692307692307\n",
      "[[ 0  3]\n",
      " [ 0 19]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.00      0.00      0.00         3\n",
      "           1       0.86      1.00      0.93        19\n",
      "\n",
      "    accuracy                           0.86        22\n",
      "   macro avg       0.43      0.50      0.46        22\n",
      "weighted avg       0.75      0.86      0.80        22\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import cross_val_score\n",
    "# tune regularization for multinomial logistic regression\n",
    "from numpy import mean\n",
    "from numpy import std\n",
    "\n",
    "# Multinomial Logistic Regression: Modified version of logistic regression that predicts a multinomial probability \n",
    "#(i.e. more than two classes) for each input example.\n",
    "\n",
    "# The LogisticRegression class can be configured for multinomial logistic regression by setting the “multi_class” \n",
    "#argument to “multinomial” and the “solver” argument to a solver that supports multinomial logistic regression, \n",
    "#such as “lbfgs“.\n",
    "\n",
    "# define the multinomial logistic regression model\n",
    "clf = LogisticRegression(multi_class=\"multinomial\", solver='lbfgs')\n",
    "\n",
    "#The multinomial logistic regression model will be fit using cross-entropy \n",
    "#loss and will predict the integer value for each integer encoded class label.\n",
    "\n",
    "# fit the model on the whole dataset\n",
    "clf.fit(scaledTrain_df, y_train.values.ravel())\n",
    "\n",
    "print('Mean accuracy in the training:', clf.score(scaledTrain_df, y_train.values.ravel())) #Apply transforms, and score with the final estimator\n",
    "\n",
    "y_pred = clf.predict(scaledValidation_df)\n",
    "\n",
    "print(confusion_matrix(y_test,y_pred))\n",
    "print(classification_report(y_test,y_pred)) #Output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature: 0, Score: -0.22142\n",
      "Feature: 1, Score: -0.05205\n",
      "Feature: 2, Score: -0.15903\n",
      "Feature: 3, Score: -0.05871\n",
      "Feature: 4, Score: 0.14778\n",
      "Feature: 5, Score: 0.00696\n",
      "Feature: 6, Score: 0.08287\n",
      "Feature: 7, Score: -0.02312\n",
      "Feature: 8, Score: -0.00016\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get importance\n",
    "importance = clf.coef_[0]\n",
    "# summarize feature importance\n",
    "for i,v in enumerate(importance):\n",
    "\tprint('Feature: %0d, Score: %.5f' % (i,v))\n",
    "# plot feature importance\n",
    "plt.bar([x for x in range(len(importance))], importance)\n",
    "plt.xticks(ticks=range(len(importance)),labels=list(scaledTrain_df.columns),rotation=90)\n",
    "plt.title(\"Importance of parameters\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finish"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### I used randome forest model and use GINI score as importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kejahanmir\\Anaconda3\\envs\\my-rdkit-env\\lib\\site-packages\\ipykernel_launcher.py:7: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  import sys\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(n_estimators=500)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Build random forest model\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=500)\n",
    "\n",
    "rf.fit(scaledTrain_df, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred = rf.predict(scaledTrain_df)\n",
    "y_test_pred = rf.predict(scaledValidation_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8636363636363636"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(scaledValidation_df,y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### different ways of calculating  Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8636363636363636"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "accuracy_score(y_test, y_test_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import balanced_accuracy_score     #because the dataset is very imbalanced    balance accuracy is better to used as metrics\n",
    "balanced_accuracy_score(y_test, y_test_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0,  3],\n",
       "       [ 0, 19]], dtype=int64)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Confusion matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "confusion_matrix(y_test, y_test_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Classification report\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "model_report = classification_report(y_train, y_train_pred, target_names=['positive','negative'])\n",
    "\n",
    "f = open('model_report.txt','w')\n",
    "f.writelines(model_report) \n",
    "f.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC curve\n",
    "# ROC curve    for sklearn from version 1.2 or newer\n",
    "from sklearn.metrics import RocCurveDisplay\n",
    "rf_disp_test = RocCurveDisplay.from_estimator(rf, scaledValidation_df, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  feature importance from the RF model\n",
    "importance = pd.Series(rf.feature_importances_, name = 'Gini')\n",
    "\n",
    "#  feature names\n",
    "feature_names = pd.Series(scaledTrain_df.columns, name = 'Feature')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Feature</th>\n",
       "      <th>Gini</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dose</td>\n",
       "      <td>0.099817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>P_density</td>\n",
       "      <td>0.337234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CSKKKK</td>\n",
       "      <td>0.113772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>QHREDGS</td>\n",
       "      <td>0.081551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>RKDVY</td>\n",
       "      <td>0.072016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>WKYMVM</td>\n",
       "      <td>0.081387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CLPFFD</td>\n",
       "      <td>0.115898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>QEINSSY</td>\n",
       "      <td>0.098326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Size_HNP</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Feature      Gini\n",
       "0       Dose  0.099817\n",
       "1  P_density  0.337234\n",
       "2     CSKKKK  0.113772\n",
       "3    QHREDGS  0.081551\n",
       "4      RKDVY  0.072016\n",
       "5     WKYMVM  0.081387\n",
       "6     CLPFFD  0.115898\n",
       "7    QEINSSY  0.098326\n",
       "8   Size_HNP  0.000000"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Combine feature names and Gini values into a Dataframe\n",
    "df = pd.concat([feature_names, importance], axis=1, names=['Feature', 'Gini'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Feature Importance')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot of feature importance\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "df_sorted = df.sort_values('Gini', ascending=False)[:20] # Sort by Gini in descending order; Showing only the top 20 results\n",
    "\n",
    "plt.figure(figsize=(5, 10))\n",
    "sns.set_theme(style=\"whitegrid\")\n",
    "ax = sns.barplot(x = 'Gini', y = 'Feature', data = df_sorted)\n",
    "plt.xlabel(\"Feature Importance\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}