06RFEforOpt_descriptors.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "937e5dfc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "import numpy as np\n",
    "from numpy import mean\n",
    "from numpy import std\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import RepeatedKFold\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.pipeline import Pipeline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "52e63673",
   "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>ID</th>\n",
       "      <th>TC</th>\n",
       "      <th>Monomer_length</th>\n",
       "      <th>MW_ratio</th>\n",
       "      <th>Mass_max</th>\n",
       "      <th>Mass_ave</th>\n",
       "      <th>K_bond_min</th>\n",
       "      <th>K_bond_ave</th>\n",
       "      <th>K_ang_ave</th>\n",
       "      <th>Theta0_min</th>\n",
       "      <th>...</th>\n",
       "      <th>Mor19</th>\n",
       "      <th>Mor31</th>\n",
       "      <th>SMR_VSA1</th>\n",
       "      <th>SMR_VSA3</th>\n",
       "      <th>SlogP_VSA5</th>\n",
       "      <th>SlogP_VSA10</th>\n",
       "      <th>EState_VSA1</th>\n",
       "      <th>EState_VSA5</th>\n",
       "      <th>VSA_EState4</th>\n",
       "      <th>VSA_EState7</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PI1</td>\n",
       "      <td>0.456185</td>\n",
       "      <td>2.58</td>\n",
       "      <td>1</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>218.51</td>\n",
       "      <td>325.817</td>\n",
       "      <td>48.657</td>\n",
       "      <td>88.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.205252</td>\n",
       "      <td>0.101002</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PI2</td>\n",
       "      <td>0.258773</td>\n",
       "      <td>2.6</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>218.51</td>\n",
       "      <td>326.68</td>\n",
       "      <td>49.1032</td>\n",
       "      <td>88.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.421298</td>\n",
       "      <td>0.185377</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.92374</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.42082</td>\n",
       "      <td>0</td>\n",
       "      <td>1.24537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PI3</td>\n",
       "      <td>0.264363</td>\n",
       "      <td>2.65</td>\n",
       "      <td>0.5</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>218.51</td>\n",
       "      <td>327.04</td>\n",
       "      <td>48.9532</td>\n",
       "      <td>88.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.350261</td>\n",
       "      <td>0.245392</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.3446</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12.8416</td>\n",
       "      <td>0</td>\n",
       "      <td>2.79977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PI4</td>\n",
       "      <td>0.261131</td>\n",
       "      <td>2.59</td>\n",
       "      <td>0.4</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>218.51</td>\n",
       "      <td>327.256</td>\n",
       "      <td>48.8663</td>\n",
       "      <td>88.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.823443</td>\n",
       "      <td>0.316596</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19.7654</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12.8416</td>\n",
       "      <td>0</td>\n",
       "      <td>4.32787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PI5</td>\n",
       "      <td>0.234342</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.4</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>218.51</td>\n",
       "      <td>327.256</td>\n",
       "      <td>49.2038</td>\n",
       "      <td>88.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.819376</td>\n",
       "      <td>0.293415</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.8475</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.42082</td>\n",
       "      <td>0</td>\n",
       "      <td>1.26389</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",
       "      <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>1046</th>\n",
       "      <td>PI1134</td>\n",
       "      <td>0.239876</td>\n",
       "      <td>19.74</td>\n",
       "      <td>0.745552</td>\n",
       "      <td>18.998</td>\n",
       "      <td>9.71751</td>\n",
       "      <td>232.52</td>\n",
       "      <td>372.619</td>\n",
       "      <td>69.222</td>\n",
       "      <td>104.88</td>\n",
       "      <td>...</td>\n",
       "      <td>0.199954</td>\n",
       "      <td>0.654512</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>19.9359</td>\n",
       "      <td>11.1269</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>28.8945</td>\n",
       "      <td>24.2655</td>\n",
       "      <td>-0.487784</td>\n",
       "      <td>-10.1832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1047</th>\n",
       "      <td>PI1135</td>\n",
       "      <td>0.248861</td>\n",
       "      <td>19.43</td>\n",
       "      <td>0.745552</td>\n",
       "      <td>18.998</td>\n",
       "      <td>9.71751</td>\n",
       "      <td>232.52</td>\n",
       "      <td>372.619</td>\n",
       "      <td>69.222</td>\n",
       "      <td>104.88</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0888365</td>\n",
       "      <td>0.626717</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>19.9359</td>\n",
       "      <td>11.1269</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>28.8945</td>\n",
       "      <td>24.2655</td>\n",
       "      <td>-0.466391</td>\n",
       "      <td>-10.182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048</th>\n",
       "      <td>PI1136</td>\n",
       "      <td>0.208098</td>\n",
       "      <td>15.11</td>\n",
       "      <td>1</td>\n",
       "      <td>32.067</td>\n",
       "      <td>8.37947</td>\n",
       "      <td>209.78</td>\n",
       "      <td>377.212</td>\n",
       "      <td>62.7439</td>\n",
       "      <td>102.44</td>\n",
       "      <td>...</td>\n",
       "      <td>1.14708</td>\n",
       "      <td>0.788496</td>\n",
       "      <td>16.8356</td>\n",
       "      <td>9.44419</td>\n",
       "      <td>25.6833</td>\n",
       "      <td>0</td>\n",
       "      <td>20.9135</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.32871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1049</th>\n",
       "      <td>PI1137</td>\n",
       "      <td>0.199093</td>\n",
       "      <td>5.34</td>\n",
       "      <td>1</td>\n",
       "      <td>32.067</td>\n",
       "      <td>10.7818</td>\n",
       "      <td>209.78</td>\n",
       "      <td>398.584</td>\n",
       "      <td>67.9282</td>\n",
       "      <td>104.44</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108499</td>\n",
       "      <td>0.326322</td>\n",
       "      <td>8.4178</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10.7042</td>\n",
       "      <td>6.06637</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1050</th>\n",
       "      <td>PI1138</td>\n",
       "      <td>0.226688</td>\n",
       "      <td>5.03</td>\n",
       "      <td>0.750083</td>\n",
       "      <td>15.999</td>\n",
       "      <td>7.06771</td>\n",
       "      <td>250.32</td>\n",
       "      <td>367.916</td>\n",
       "      <td>57.3668</td>\n",
       "      <td>107.58</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0672052</td>\n",
       "      <td>0.086131</td>\n",
       "      <td>5.10653</td>\n",
       "      <td>0</td>\n",
       "      <td>11.1269</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>24.117</td>\n",
       "      <td>2.77231</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1051 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          ID        TC Monomer_length  MW_ratio Mass_max Mass_ave K_bond_min  \\\n",
       "0        PI1  0.456185           2.58         1   12.011  4.67567     218.51   \n",
       "1        PI2  0.258773            2.6  0.666667   12.011  4.67567     218.51   \n",
       "2        PI3  0.264363           2.65       0.5   12.011  4.67567     218.51   \n",
       "3        PI4  0.261131           2.59       0.4   12.011  4.67567     218.51   \n",
       "4        PI5  0.234342           2.76       0.4   12.011  4.67567     218.51   \n",
       "...      ...       ...            ...       ...      ...      ...        ...   \n",
       "1046  PI1134  0.239876          19.74  0.745552   18.998  9.71751     232.52   \n",
       "1047  PI1135  0.248861          19.43  0.745552   18.998  9.71751     232.52   \n",
       "1048  PI1136  0.208098          15.11         1   32.067  8.37947     209.78   \n",
       "1049  PI1137  0.199093           5.34         1   32.067  10.7818     209.78   \n",
       "1050  PI1138  0.226688           5.03  0.750083   15.999  7.06771     250.32   \n",
       "\n",
       "     K_bond_ave K_ang_ave Theta0_min  ...      Mor19     Mor31 SMR_VSA1  \\\n",
       "0       325.817    48.657       88.4  ...   0.205252  0.101002        0   \n",
       "1        326.68   49.1032       88.4  ...   0.421298  0.185377        0   \n",
       "2        327.04   48.9532       88.4  ...   0.350261  0.245392        0   \n",
       "3       327.256   48.8663       88.4  ...   0.823443  0.316596        0   \n",
       "4       327.256   49.2038       88.4  ...   0.819376  0.293415        0   \n",
       "...         ...       ...        ...  ...        ...       ...      ...   \n",
       "1046    372.619    69.222     104.88  ...   0.199954  0.654512  26.3425   \n",
       "1047    372.619    69.222     104.88  ...  0.0888365  0.626717  26.3425   \n",
       "1048    377.212   62.7439     102.44  ...    1.14708  0.788496  16.8356   \n",
       "1049    398.584   67.9282     104.44  ...  -0.108499  0.326322   8.4178   \n",
       "1050    367.916   57.3668     107.58  ...  0.0672052  0.086131  5.10653   \n",
       "\n",
       "     SMR_VSA3 SlogP_VSA5 SlogP_VSA10 EState_VSA1 EState_VSA5 VSA_EState4  \\\n",
       "0           0          0           0           0           0           0   \n",
       "1           0    6.92374           0           0     6.42082           0   \n",
       "2           0    13.3446           0           0     12.8416           0   \n",
       "3           0    19.7654           0           0     12.8416           0   \n",
       "4           0    13.8475           0           0     6.42082           0   \n",
       "...       ...        ...         ...         ...         ...         ...   \n",
       "1046  19.9359    11.1269     26.3425     28.8945     24.2655   -0.487784   \n",
       "1047  19.9359    11.1269     26.3425     28.8945     24.2655   -0.466391   \n",
       "1048  9.44419    25.6833           0     20.9135           0           0   \n",
       "1049        0          0           0     10.7042     6.06637           0   \n",
       "1050        0    11.1269           0           0      24.117     2.77231   \n",
       "\n",
       "     VSA_EState7  \n",
       "0              0  \n",
       "1        1.24537  \n",
       "2        2.79977  \n",
       "3        4.32787  \n",
       "4        1.26389  \n",
       "...          ...  \n",
       "1046    -10.1832  \n",
       "1047     -10.182  \n",
       "1048     3.32871  \n",
       "1049           0  \n",
       "1050           0  \n",
       "\n",
       "[1051 rows x 55 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path=\"./dataset/\"\n",
    "file='Cor_descriptor.pkl'\n",
    "df= pd.read_pickle(path+file)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "5861e3ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_all=df.drop(df.columns[0:2], axis=1)\n",
    "y_all=df[\"TC\"]\n",
    "for i in [X_all,y_all]:\n",
    "    i.index = range(i.shape[0])\n",
    "scaler.fit(X_all)\n",
    "X = scaler.transform(X_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bf4b1fca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "411 16 4 4\n"
     ]
    }
   ],
   "source": [
    "#Bayesian optimized hyperparameters\n",
    "n_estimator = int(411.07423627934196)\n",
    "max_depths = int(16.808976918640724)\n",
    "min_samples_split=int(np.round(3.530568310298654))\n",
    "min_samples_leaf=int(np.round(3.7540007963006756))\n",
    "print(n_estimator,max_depths,min_samples_split,min_samples_leaf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2cb6840e",
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestRegressor(n_estimators = n_estimator, max_features='auto',random_state=1, max_depth = max_depths,min_samples_split=min_samples_split,min_samples_leaf=min_samples_leaf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6f5bdcbb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_models():\n",
    "    models = dict()\n",
    "    for i in range(10, 54):\n",
    "        rfe = RFE(estimator=rf, n_features_to_select=i)\n",
    "        model =rf\n",
    "        models[str(i)] = Pipeline(steps=[('s',rfe),('m',model)])\n",
    "    return models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "488d454b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# evaluate a give model using cross-validation\n",
    "def evaluate_model(model, X, y):\n",
    "    cv =RepeatedKFold(n_splits=10, n_repeats=1,random_state=1)\n",
    "    scores = cross_val_score(model,X,y, scoring='r2', cv=cv, n_jobs=-1, error_score='raise')\n",
    "    scores_mse=cross_val_score(model,X,y, scoring='neg_mean_squared_error', cv=cv, n_jobs=-1, error_score='raise')\n",
    "    return scores,scores_mse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0d5bf5ec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">10 0.617 (0.137)\n",
      ">11 0.617 (0.136)\n",
      ">12 0.618 (0.137)\n",
      ">13 0.621 (0.140)\n",
      ">14 0.627 (0.134)\n",
      ">15 0.632 (0.135)\n",
      ">16 0.636 (0.133)\n",
      ">17 0.637 (0.134)\n",
      ">18 0.636 (0.131)\n",
      ">19 0.637 (0.131)\n",
      ">20 0.638 (0.132)\n",
      ">21 0.640 (0.131)\n",
      ">22 0.643 (0.131)\n",
      ">23 0.643 (0.133)\n",
      ">24 0.646 (0.134)\n",
      ">25 0.651 (0.133)\n",
      ">26 0.654 (0.130)\n",
      ">27 0.655 (0.132)\n",
      ">28 0.655 (0.133)\n",
      ">29 0.657 (0.133)\n",
      ">30 0.656 (0.133)\n",
      ">31 0.658 (0.132)\n",
      ">32 0.657 (0.132)\n",
      ">33 0.657 (0.133)\n",
      ">34 0.658 (0.134)\n",
      ">35 0.659 (0.134)\n",
      ">36 0.659 (0.135)\n",
      ">37 0.661 (0.135)\n",
      ">38 0.661 (0.134)\n",
      ">39 0.662 (0.134)\n",
      ">40 0.663 (0.133)\n",
      ">41 0.664 (0.133)\n",
      ">42 0.664 (0.133)\n",
      ">43 0.664 (0.133)\n",
      ">44 0.663 (0.133)\n",
      ">45 0.664 (0.133)\n",
      ">46 0.664 (0.133)\n",
      ">47 0.664 (0.133)\n",
      ">48 0.664 (0.133)\n",
      ">49 0.664 (0.132)\n",
      ">50 0.664 (0.133)\n",
      ">51 0.664 (0.133)\n",
      ">52 0.665 (0.133)\n",
      ">53 0.664 (0.133)\n"
     ]
    }
   ],
   "source": [
    "models = get_models()\n",
    "# evaluate the models and store results\n",
    "results,MSE,names = list(), list(), list()\n",
    "for name, model in models.items():\n",
    "    scores,scores_mse = evaluate_model(model, X, y_all)\n",
    "    results.append(scores)\n",
    "    MSE.append(scores)\n",
    "    names.append(scores_mse)\n",
    "    print('>%s %.3f (%.3f)' % (name, mean(scores), std(scores)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "eb486dfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "R2_MEAN=[]\n",
    "R2_STD=[]\n",
    "MSE_MEAN=[]\n",
    "MSE_STD=[]\n",
    "random_Number=[]\n",
    "for i in range (0,len(results)):\n",
    "    R2_MEAN.append(mean(results[i]))\n",
    "    R2_STD.append(std(results[i])/2)\n",
    "    MSE_MEAN.append(-1*mean(names[i]))\n",
    "    MSE_STD.append(std(names[i])/2)\n",
    "    random_Number.append(i+10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "c3ffa17e",
   "metadata": {},
   "outputs": [],
   "source": [
    "R2_MEAN=np.array(R2_MEAN)\n",
    "R2_STD=np.array(R2_STD)\n",
    "random_Number=np.array(random_Number)\n",
    "MSE_MEAN=np.array(MSE_MEAN)\n",
    "MSE_STD=np.array(MSE_STD)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "39a9cf03",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\huang\\AppData\\Local\\Temp/ipykernel_69348/125263462.py:27: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  axes = plt.subplot()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 708.661x472.441 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "inch2cm=1/2.54\n",
    "plt.figure(figsize=(6*inch2cm, 4*inch2cm))\n",
    "plt.rcParams[\"figure.dpi\"] = 300  # 分辨率\n",
    "plt.rc('font',family='Arial',weight='normal')\n",
    "font1 = {'family' : 'Arial',\n",
    "'weight' : 'normal',\n",
    "'size'   : 10,\n",
    "}\n",
    "\n",
    "plt.plot(random_Number, R2_MEAN, '-', color='k',label='Random')\n",
    "plt.fill_between(random_Number,R2_MEAN-R2_STD,R2_MEAN+R2_STD,\n",
    "                 color='k', alpha=0.2)\n",
    "\n",
    "plt.xlim(10,53)\n",
    "plt.ylim(0.2,0.8)\n",
    "plt.xticks(fontsize=10) \n",
    "plt.yticks(fontsize=10)\n",
    "my_x_ticks = np.arange(10,53,5)\n",
    "my_y_ticks = np.arange(0.2,0.81,0.1)\n",
    "plt.xticks(my_x_ticks,size=12)\n",
    "plt.yticks(my_y_ticks,size=12)\n",
    "plt.grid(False)\n",
    "axes = plt.subplot()\n",
    "axes.minorticks_on()\n",
    "axes.tick_params(axis=\"both\", which=\"major\", direction=\"out\", width=1.5, length=3)\n",
    "TK = plt.gca()\n",
    "bwith = 1.5\n",
    "TK.spines['bottom'].set_linewidth(bwith)\n",
    "TK.spines['left'].set_linewidth(bwith)\n",
    "TK.spines['top'].set_linewidth(bwith)\n",
    "TK.spines['right'].set_linewidth(bwith)\n",
    "TK.spines['bottom'].set_color('k')\n",
    "TK.spines['left'].set_color('k')\n",
    "TK.spines['top'].set_color('k')\n",
    "TK.spines['right'].set_color('k')\n",
    "plt.xlabel(\"Number of descriptors\",fontsize=10)\n",
    "plt.ylabel(\"R2\",fontsize=10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "f2339731",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\huang\\AppData\\Local\\Temp/ipykernel_69348/2868686199.py:27: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  axes = plt.subplot()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 708.661x472.441 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "inch2cm=1/2.54\n",
    "plt.figure(figsize=(6*inch2cm, 4*inch2cm))\n",
    "plt.rcParams[\"figure.dpi\"] = 300  # 分辨率\n",
    "plt.rc('font',family='Arial',weight='normal')\n",
    "font1 = {'family' : 'Arial',\n",
    "'weight' : 'normal',\n",
    "'size'   : 10,\n",
    "}\n",
    "plt.plot(random_Number, MSE_MEAN, '-', color='k',label='Random')\n",
    "plt.fill_between(random_Number,MSE_MEAN-MSE_STD,MSE_MEAN+MSE_STD,\n",
    "                 color='k', alpha=0.2)\n",
    "plt.xlim(10,53)\n",
    "plt.ylim(0,0.002)\n",
    "plt.xticks(fontsize=10)\n",
    "plt.yticks(fontsize=10)\n",
    "my_x_ticks = np.arange(10,53,5)\n",
    "my_y_ticks = np.arange(0,0.003,0.0005)\n",
    "plt.xticks(my_x_ticks,size=12)\n",
    "plt.yticks(my_y_ticks,size=12)\n",
    "plt.grid(False)\n",
    "axes = plt.subplot()\n",
    "axes.minorticks_on()\n",
    "axes.tick_params(axis=\"both\", which=\"major\", direction=\"out\", width=1.5, length=3)\n",
    "TK = plt.gca()\n",
    "bwith = 1.5\n",
    "TK.spines['bottom'].set_linewidth(bwith)\n",
    "TK.spines['left'].set_linewidth(bwith)\n",
    "TK.spines['top'].set_linewidth(bwith)\n",
    "TK.spines['right'].set_linewidth(bwith)\n",
    "TK.spines['bottom'].set_color('k')\n",
    "TK.spines['left'].set_color('k')\n",
    "TK.spines['top'].set_color('k')\n",
    "TK.spines['right'].set_color('k')\n",
    "plt.xlabel(\"Number of descriptors\",fontsize=10)\n",
    "plt.ylabel(\"MSE\",fontsize=10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "4b538361",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE: -0.001 (0.001)\n",
      "R2: 0.664 (0.133)\n"
     ]
    }
   ],
   "source": [
    "rfe = RFE(estimator=rf, n_features_to_select=25)\n",
    "model = rf\n",
    "pipeline = Pipeline(steps=[('s',rfe),('m',model)])\n",
    "# evaluate model\n",
    "cv =RepeatedKFold(n_splits=10, n_repeats=1,random_state=1)\n",
    "scores = cross_val_score(model, X, y_all, scoring='r2', cv=cv, n_jobs=-1, error_score='raise')\n",
    "scores_mse=cross_val_score(model,  X, y_all,  scoring='neg_mean_squared_error', cv=cv, n_jobs=-1, error_score='raise')\n",
    "# report performance\n",
    "print('MSE: %.3f (%.3f)' % (mean(scores_mse), std(scores_mse)))\n",
    "print('R2: %.3f (%.3f)' % (mean(scores), std(scores)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "66c690a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "selector =rfe.fit(X, y_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "bf57b520",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Monomer_length', 'MW_ratio', 'Mass_max', 'Mass_ave', 'K_bond_min',\n",
       "       'K_bond_ave', 'K_ang_ave', 'Theta0_min', 'nAtom', 'nH', 'nN', 'nF',\n",
       "       'nCl', 'AATS0dv', 'AATS0d', 'AATS0Z', 'AATS1Z', 'ATSC2Z', 'ATSC5Z',\n",
       "       'AATSC0dv', 'AATSC0Z', 'AATSC2Z', 'BCUTdv-1h', 'BCUTd-1h', 'BCUTZ-1h',\n",
       "       'C1SP2', 'Xc-3d', 'Xc-5d', 'Xc-6d', 'NssssC', 'ETA_eta_B', 'nHBDon',\n",
       "       'CIC0', 'CIC1', 'MIC0', 'MIC1', 'Kier2', 'Kier3', 'Mor02', 'Mor04',\n",
       "       'Mor08', 'Mor13', 'Mor14', 'Mor19', 'Mor31', 'SMR_VSA1', 'SMR_VSA3',\n",
       "       'SlogP_VSA5', 'SlogP_VSA10', 'EState_VSA1', 'EState_VSA5',\n",
       "       'VSA_EState4', 'VSA_EState7'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sele=selector.ranking_\n",
    "feature_name=X_all.columns\n",
    "feature_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "6bc91f32",
   "metadata": {},
   "outputs": [],
   "source": [
    "Opt_desc=[]\n",
    "for i in range (0,len(sele)):\n",
    "    if sele[i]==1:\n",
    "        Opt_desc.append(feature_name[i])\n",
    "df1=df[Opt_desc]\n",
    "to_save=df[[\"ID\",\"TC\"]].join(df1)\n",
    "file1 = r\"Opt_descriptor.pkl\"\n",
    "to_save.to_pickle(path+file1)        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "f55f97d8",
   "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>ID</th>\n",
       "      <th>TC</th>\n",
       "      <th>Monomer_length</th>\n",
       "      <th>MW_ratio</th>\n",
       "      <th>Mass_max</th>\n",
       "      <th>Mass_ave</th>\n",
       "      <th>K_bond_ave</th>\n",
       "      <th>K_ang_ave</th>\n",
       "      <th>nH</th>\n",
       "      <th>AATS0d</th>\n",
       "      <th>...</th>\n",
       "      <th>nHBDon</th>\n",
       "      <th>MIC0</th>\n",
       "      <th>MIC1</th>\n",
       "      <th>Kier3</th>\n",
       "      <th>Mor02</th>\n",
       "      <th>Mor14</th>\n",
       "      <th>Mor19</th>\n",
       "      <th>SMR_VSA1</th>\n",
       "      <th>SMR_VSA3</th>\n",
       "      <th>VSA_EState4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PI1</td>\n",
       "      <td>0.456185</td>\n",
       "      <td>2.58</td>\n",
       "      <td>1</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>325.817</td>\n",
       "      <td>48.657</td>\n",
       "      <td>4</td>\n",
       "      <td>1.75</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5095</td>\n",
       "      <td>6.5095</td>\n",
       "      <td>4</td>\n",
       "      <td>13.5611</td>\n",
       "      <td>0.233769</td>\n",
       "      <td>0.205252</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PI2</td>\n",
       "      <td>0.258773</td>\n",
       "      <td>2.6</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>326.68</td>\n",
       "      <td>49.1032</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6.62105</td>\n",
       "      <td>11.813</td>\n",
       "      <td>4</td>\n",
       "      <td>19.2292</td>\n",
       "      <td>0.393246</td>\n",
       "      <td>0.421298</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PI3</td>\n",
       "      <td>0.264363</td>\n",
       "      <td>2.65</td>\n",
       "      <td>0.5</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>327.04</td>\n",
       "      <td>48.9532</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6.66736</td>\n",
       "      <td>13.5308</td>\n",
       "      <td>3</td>\n",
       "      <td>21.9293</td>\n",
       "      <td>0.675759</td>\n",
       "      <td>0.350261</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PI4</td>\n",
       "      <td>0.261131</td>\n",
       "      <td>2.59</td>\n",
       "      <td>0.4</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>327.256</td>\n",
       "      <td>48.8663</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6.69093</td>\n",
       "      <td>13.4804</td>\n",
       "      <td>3.84</td>\n",
       "      <td>24.7537</td>\n",
       "      <td>0.923227</td>\n",
       "      <td>0.823443</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PI5</td>\n",
       "      <td>0.234342</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.4</td>\n",
       "      <td>12.011</td>\n",
       "      <td>4.67567</td>\n",
       "      <td>327.256</td>\n",
       "      <td>49.2038</td>\n",
       "      <td>10</td>\n",
       "      <td>2.11765</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6.69093</td>\n",
       "      <td>13.4804</td>\n",
       "      <td>2.66667</td>\n",
       "      <td>25.6625</td>\n",
       "      <td>0.872534</td>\n",
       "      <td>0.819376</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "      <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>1046</th>\n",
       "      <td>PI1134</td>\n",
       "      <td>0.239876</td>\n",
       "      <td>19.74</td>\n",
       "      <td>0.745552</td>\n",
       "      <td>18.998</td>\n",
       "      <td>9.71751</td>\n",
       "      <td>372.619</td>\n",
       "      <td>69.222</td>\n",
       "      <td>16</td>\n",
       "      <td>4.42105</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>16.7388</td>\n",
       "      <td>33.6424</td>\n",
       "      <td>5.776</td>\n",
       "      <td>43.1547</td>\n",
       "      <td>-0.537266</td>\n",
       "      <td>0.199954</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>19.9359</td>\n",
       "      <td>-0.487784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1047</th>\n",
       "      <td>PI1135</td>\n",
       "      <td>0.248861</td>\n",
       "      <td>19.43</td>\n",
       "      <td>0.745552</td>\n",
       "      <td>18.998</td>\n",
       "      <td>9.71751</td>\n",
       "      <td>372.619</td>\n",
       "      <td>69.222</td>\n",
       "      <td>16</td>\n",
       "      <td>4.42105</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>16.7388</td>\n",
       "      <td>33.6424</td>\n",
       "      <td>5.776</td>\n",
       "      <td>44.1209</td>\n",
       "      <td>-0.00156469</td>\n",
       "      <td>0.0888365</td>\n",
       "      <td>26.3425</td>\n",
       "      <td>19.9359</td>\n",
       "      <td>-0.466391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048</th>\n",
       "      <td>PI1136</td>\n",
       "      <td>0.208098</td>\n",
       "      <td>15.11</td>\n",
       "      <td>1</td>\n",
       "      <td>32.067</td>\n",
       "      <td>8.37947</td>\n",
       "      <td>377.212</td>\n",
       "      <td>62.7439</td>\n",
       "      <td>18</td>\n",
       "      <td>3.05</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>22.0525</td>\n",
       "      <td>31.724</td>\n",
       "      <td>9.03686</td>\n",
       "      <td>44.7162</td>\n",
       "      <td>1.69004</td>\n",
       "      <td>1.14708</td>\n",
       "      <td>16.8356</td>\n",
       "      <td>9.44419</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1049</th>\n",
       "      <td>PI1137</td>\n",
       "      <td>0.199093</td>\n",
       "      <td>5.34</td>\n",
       "      <td>1</td>\n",
       "      <td>32.067</td>\n",
       "      <td>10.7818</td>\n",
       "      <td>398.584</td>\n",
       "      <td>67.9282</td>\n",
       "      <td>4</td>\n",
       "      <td>3.86667</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>21.4168</td>\n",
       "      <td>27.4301</td>\n",
       "      <td>2.5</td>\n",
       "      <td>13.9872</td>\n",
       "      <td>-0.13399</td>\n",
       "      <td>-0.108499</td>\n",
       "      <td>8.4178</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1050</th>\n",
       "      <td>PI1138</td>\n",
       "      <td>0.226688</td>\n",
       "      <td>5.03</td>\n",
       "      <td>0.750083</td>\n",
       "      <td>15.999</td>\n",
       "      <td>7.06771</td>\n",
       "      <td>367.916</td>\n",
       "      <td>57.3668</td>\n",
       "      <td>8</td>\n",
       "      <td>3.15789</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>10.4177</td>\n",
       "      <td>23.6121</td>\n",
       "      <td>1.97531</td>\n",
       "      <td>17.8576</td>\n",
       "      <td>0.300494</td>\n",
       "      <td>0.0672052</td>\n",
       "      <td>5.10653</td>\n",
       "      <td>0</td>\n",
       "      <td>2.77231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1051 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          ID        TC Monomer_length  MW_ratio Mass_max Mass_ave K_bond_ave  \\\n",
       "0        PI1  0.456185           2.58         1   12.011  4.67567    325.817   \n",
       "1        PI2  0.258773            2.6  0.666667   12.011  4.67567     326.68   \n",
       "2        PI3  0.264363           2.65       0.5   12.011  4.67567     327.04   \n",
       "3        PI4  0.261131           2.59       0.4   12.011  4.67567    327.256   \n",
       "4        PI5  0.234342           2.76       0.4   12.011  4.67567    327.256   \n",
       "...      ...       ...            ...       ...      ...      ...        ...   \n",
       "1046  PI1134  0.239876          19.74  0.745552   18.998  9.71751    372.619   \n",
       "1047  PI1135  0.248861          19.43  0.745552   18.998  9.71751    372.619   \n",
       "1048  PI1136  0.208098          15.11         1   32.067  8.37947    377.212   \n",
       "1049  PI1137  0.199093           5.34         1   32.067  10.7818    398.584   \n",
       "1050  PI1138  0.226688           5.03  0.750083   15.999  7.06771    367.916   \n",
       "\n",
       "     K_ang_ave  nH   AATS0d  ... nHBDon     MIC0     MIC1    Kier3    Mor02  \\\n",
       "0       48.657   4     1.75  ...      0   6.5095   6.5095        4  13.5611   \n",
       "1      49.1032   6        2  ...      0  6.62105   11.813        4  19.2292   \n",
       "2      48.9532   8        2  ...      0  6.66736  13.5308        3  21.9293   \n",
       "3      48.8663  10        2  ...      0  6.69093  13.4804     3.84  24.7537   \n",
       "4      49.2038  10  2.11765  ...      0  6.69093  13.4804  2.66667  25.6625   \n",
       "...        ...  ..      ...  ...    ...      ...      ...      ...      ...   \n",
       "1046    69.222  16  4.42105  ...      2  16.7388  33.6424    5.776  43.1547   \n",
       "1047    69.222  16  4.42105  ...      2  16.7388  33.6424    5.776  44.1209   \n",
       "1048   62.7439  18     3.05  ...      2  22.0525   31.724  9.03686  44.7162   \n",
       "1049   67.9282   4  3.86667  ...      0  21.4168  27.4301      2.5  13.9872   \n",
       "1050   57.3668   8  3.15789  ...      1  10.4177  23.6121  1.97531  17.8576   \n",
       "\n",
       "           Mor14      Mor19 SMR_VSA1 SMR_VSA3 VSA_EState4  \n",
       "0       0.233769   0.205252        0        0           0  \n",
       "1       0.393246   0.421298        0        0           0  \n",
       "2       0.675759   0.350261        0        0           0  \n",
       "3       0.923227   0.823443        0        0           0  \n",
       "4       0.872534   0.819376        0        0           0  \n",
       "...          ...        ...      ...      ...         ...  \n",
       "1046   -0.537266   0.199954  26.3425  19.9359   -0.487784  \n",
       "1047 -0.00156469  0.0888365  26.3425  19.9359   -0.466391  \n",
       "1048     1.69004    1.14708  16.8356  9.44419           0  \n",
       "1049    -0.13399  -0.108499   8.4178        0           0  \n",
       "1050    0.300494  0.0672052  5.10653        0     2.77231  \n",
       "\n",
       "[1051 rows x 27 columns]"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "to_save"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}