6.1 - PhOLED 300 KERAS.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(313, 32)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>homo_HIL</th>\n",
       "      <th>thickness_HIL</th>\n",
       "      <th>lumo_HTL1</th>\n",
       "      <th>homo_HTL1</th>\n",
       "      <th>thickness_HTL1</th>\n",
       "      <th>lumo_HTL2</th>\n",
       "      <th>homo_HTL2</th>\n",
       "      <th>thickness_HTL2</th>\n",
       "      <th>lumo_HTL3</th>\n",
       "      <th>homo_HTL3</th>\n",
       "      <th>...</th>\n",
       "      <th>total_ETL_thickness</th>\n",
       "      <th>lumo_ETL1</th>\n",
       "      <th>homo_ETL1</th>\n",
       "      <th>thickness_ETL1</th>\n",
       "      <th>lumo_ETL2</th>\n",
       "      <th>homo_ETL2</th>\n",
       "      <th>thickness_ETL2</th>\n",
       "      <th>cathode_workfunction</th>\n",
       "      <th>thickness_cathode1</th>\n",
       "      <th>cd/A</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   homo_HIL  thickness_HIL  lumo_HTL1  homo_HTL1  thickness_HTL1  lumo_HTL2  \\\n",
       "0       5.2           25.0        3.3        5.2               0        3.3   \n",
       "1       5.2           25.0        3.3        5.2               0        3.3   \n",
       "2       5.2           25.0        3.3        5.2               0        3.3   \n",
       "3       5.2           25.0        3.3        5.2               0        3.3   \n",
       "4       5.2           25.0        3.3        5.2               0        3.3   \n",
       "\n",
       "   homo_HTL2  thickness_HTL2  lumo_HTL3  homo_HTL3  ...   total_ETL_thickness  \\\n",
       "0        5.2               0        3.3        5.2  ...                  35.0   \n",
       "1        5.2               0        3.3        5.2  ...                  35.0   \n",
       "2        5.2               0        3.3        5.2  ...                  35.0   \n",
       "3        5.2               0        3.3        5.2  ...                  35.0   \n",
       "4        5.2               0        3.3        5.2  ...                  35.0   \n",
       "\n",
       "   lumo_ETL1  homo_ETL1  thickness_ETL1  lumo_ETL2  homo_ETL2  thickness_ETL2  \\\n",
       "0        2.8        6.2               0        2.8        6.2              35   \n",
       "1        2.8        6.2               0        2.8        6.2              35   \n",
       "2        2.8        6.2               0        2.8        6.2              35   \n",
       "3        2.8        6.2               0        2.8        6.2              35   \n",
       "4        2.8        6.2               0        2.8        6.2              35   \n",
       "\n",
       "   cathode_workfunction  thickness_cathode1  cd/A  \n",
       "0                   2.8                10.0   5.0  \n",
       "1                   2.8                10.0   4.0  \n",
       "2                   2.8                10.0   9.0  \n",
       "3                   2.8                10.0  12.0  \n",
       "4                   2.8                10.0   8.0  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import KFold, LeaveOneOut\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "plt.style.use('ggplot')\n",
    "%matplotlib inline\n",
    "\n",
    "# read excel file\n",
    "df = pd.read_excel('C:/Users/acc_a/OneDrive/PhOLED 300.xlsx', sheetname='Sheet2')\n",
    "df = df[list(df.columns)[1:]]\n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(304, 32)\n"
     ]
    }
   ],
   "source": [
    "#Checking for missing data\n",
    "NAs = df.isnull().sum()\n",
    "NAs.sort_values(0, ascending=False)\n",
    "NAs[NAs > 0] #check wnumber of null values in column\n",
    "\n",
    "# np.nonzero(pd.isnull(df.values).any(1)) #check the index of missing value\n",
    "\n",
    "# drop the missing data\n",
    "df.dropna(axis=0, inplace=True)\n",
    "print(df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(304, 31)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>homo_HIL</th>\n",
       "      <th>thickness_HIL</th>\n",
       "      <th>lumo_HTL1</th>\n",
       "      <th>homo_HTL1</th>\n",
       "      <th>thickness_HTL1</th>\n",
       "      <th>lumo_HTL2</th>\n",
       "      <th>homo_HTL2</th>\n",
       "      <th>thickness_HTL2</th>\n",
       "      <th>lumo_HTL3</th>\n",
       "      <th>homo_HTL3</th>\n",
       "      <th>...</th>\n",
       "      <th>triplet_ETL</th>\n",
       "      <th>total_ETL_thickness</th>\n",
       "      <th>lumo_ETL1</th>\n",
       "      <th>homo_ETL1</th>\n",
       "      <th>thickness_ETL1</th>\n",
       "      <th>lumo_ETL2</th>\n",
       "      <th>homo_ETL2</th>\n",
       "      <th>thickness_ETL2</th>\n",
       "      <th>cathode_workfunction</th>\n",
       "      <th>thickness_cathode1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>2.6</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>2.6</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>2.6</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>2.6</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.2</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>2.6</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.2</td>\n",
       "      <td>35</td>\n",
       "      <td>2.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   homo_HIL  thickness_HIL  lumo_HTL1  homo_HTL1  thickness_HTL1  lumo_HTL2  \\\n",
       "0       5.2           25.0        3.3        5.2               0        3.3   \n",
       "1       5.2           25.0        3.3        5.2               0        3.3   \n",
       "2       5.2           25.0        3.3        5.2               0        3.3   \n",
       "3       5.2           25.0        3.3        5.2               0        3.3   \n",
       "4       5.2           25.0        3.3        5.2               0        3.3   \n",
       "\n",
       "   homo_HTL2  thickness_HTL2  lumo_HTL3  homo_HTL3         ...          \\\n",
       "0        5.2               0        3.3        5.2         ...           \n",
       "1        5.2               0        3.3        5.2         ...           \n",
       "2        5.2               0        3.3        5.2         ...           \n",
       "3        5.2               0        3.3        5.2         ...           \n",
       "4        5.2               0        3.3        5.2         ...           \n",
       "\n",
       "   triplet_ETL  total_ETL_thickness  lumo_ETL1  homo_ETL1  thickness_ETL1  \\\n",
       "0          2.6                 35.0        2.8        6.2               0   \n",
       "1          2.6                 35.0        2.8        6.2               0   \n",
       "2          2.6                 35.0        2.8        6.2               0   \n",
       "3          2.6                 35.0        2.8        6.2               0   \n",
       "4          2.6                 35.0        2.8        6.2               0   \n",
       "\n",
       "   lumo_ETL2  homo_ETL2  thickness_ETL2  cathode_workfunction  \\\n",
       "0        2.8        6.2              35                   2.8   \n",
       "1        2.8        6.2              35                   2.8   \n",
       "2        2.8        6.2              35                   2.8   \n",
       "3        2.8        6.2              35                   2.8   \n",
       "4        2.8        6.2              35                   2.8   \n",
       "\n",
       "   thickness_cathode1  \n",
       "0                10.0  \n",
       "1                10.0  \n",
       "2                10.0  \n",
       "3                10.0  \n",
       "4                10.0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.iloc[:, :-1]\n",
    "y = df[['cd/A']]\n",
    "print(X.shape)\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "rs = 24\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y,test_size=0.25, random_state=rs)\n",
    "\n",
    "kfold = KFold(n_splits=5, shuffle=True, random_state=rs)\n",
    "loo = LeaveOneOut()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras.wrappers.scikit_learn import KerasRegressor\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# define and fit the final model\n",
    "def baseline_model():\n",
    "    model = Sequential()\n",
    "    model.add(Dense(31, input_dim=X_train.shape[1], activation='relu'))\n",
    "#     model.add(Dense(31, activation='relu'))\n",
    "    model.add(Dense(1000, activation='relu'))\n",
    "    model.add(Dense(500, activation='relu'))\n",
    "    model.add(Dense(100, activation='relu'))\n",
    "    model.add(Dense(1, activation='relu'))\n",
    "    model.compile(loss='mse', optimizer='adam', metrics=['mse'])    \n",
    "    return model\n",
    "    \n",
    "regressor = KerasRegressor(build_fn=baseline_model, verbose=False)\n",
    "\n",
    "# # define the grid search parameters\n",
    "# batch_size = [5, 10]\n",
    "# epochs = [100, 250]\n",
    "# param_grid = dict(batch_size=batch_size, epochs=epochs)\n",
    "# grid = GridSearchCV(estimator=regressor, param_grid=param_grid, n_jobs=-1)\n",
    "# grid_result = grid.fit(X, y)\n",
    "\n",
    "regressor.fit(np.array(X_train), np.array(y_train));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(\"Best: %f using %s\" % (grid_result.best_score_, grid_result.best_params_))\n",
    "# means = grid_result.cv_results_['mean_test_score']\n",
    "# stds = grid_result.cv_results_['std_test_score']\n",
    "# params = grid_result.cv_results_['params']\n",
    "# for mean, stdev, param in zip(means, stds, params):\n",
    "#     print(\"%f (%f) with: %r\" % (mean, stdev, param))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predictions = regressor.predict(np.array(X_test))\n",
    "pred_train = regressor.predict(np.array(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 31.22196314  12.87124216  12.84909688  10.14965879  12.23708613]\n",
      "RMSE:  12.727587104\n"
     ]
    }
   ],
   "source": [
    "results = cross_val_score(regressor,np.array(X), np.array(y), cv=kfold)\n",
    "print(np.sqrt(results))\n",
    "print('RMSE: ', np.sqrt(mean_squared_error(y_test, y_predictions)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1b4eb76c1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (10.0, 10.0)\n",
    "plt.scatter(pred_train, y_train, c='r', s=40, label='training')\n",
    "plt.scatter(y_predictions, y_test, c='b', s=40, label='testing')\n",
    "plt.plot([0, 60], [-0,60], 'g-', label='true line')\n",
    "plt.legend(loc=4, frameon=True, title='Legend', fontsize=14);\n",
    "plt.ylabel('True efficiency', fontsize=18)\n",
    "plt.xlabel('Predicted efficiency', fontsize=18)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.yticks(fontsize=14)\n",
    "plt.title(' keras regressor ',fontsize=20)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scale data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train_norm = scaler.transform(X_train)\n",
    "X_test_norm = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "def second_model():\n",
    "    model = Sequential()\n",
    "    model.add(Dropout(0.3, input_shape=(31,)))\n",
    "#     model.add(Dense(31, input_dim=X_train.shape[1], activation='relu'))\n",
    "    model.add(Dense(1000, activation='relu'))\n",
    "#     model.add(Dropout(0.2))\n",
    "    model.add(Dense(1000, activation='relu'))\n",
    "#     model.add(Dropout(0.2))\n",
    "#     model.add(Dense(100, activation='relu'))\n",
    "#     model.add(Dropout(0.2))\n",
    "    model.add(Dense(1, activation='relu'))\n",
    "    model.compile(loss='mse', optimizer='ADAgrad', metrics=['mse'])\n",
    "    \n",
    "    return model\n",
    "\n",
    "regressor = KerasRegressor(build_fn=second_model, nb_epoch=1000000, batch_size=25, verbose=False)\n",
    "\n",
    "regressor.fit(np.array(X_train_norm), np.array(y_train))\n",
    "\n",
    "y_predictions = regressor.predict(np.array(X_test_norm))\n",
    "pred_train = regressor.predict(np.array(X_train_norm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 31.22196344  28.06738593  14.46521065  13.53543859  12.58216915]\n",
      "RMSE:  10.4892406132\n"
     ]
    }
   ],
   "source": [
    "results = cross_val_score(regressor,np.array(X), np.array(y), cv=kfold)\n",
    "print(np.sqrt(results))\n",
    "print('RMSE: ', np.sqrt(mean_squared_error(y_test, y_predictions)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1b52255d710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (10.0, 10.0)\n",
    "plt.scatter(pred_train, y_train, c='r', s=40, label='training')\n",
    "plt.scatter(y_predictions, y_test, c='b', s=40, label='testing')\n",
    "plt.plot([0, 60], [-0,60], 'g-', label='true line')\n",
    "plt.legend(loc=4, frameon=True, title='Legend', fontsize=14);\n",
    "plt.ylabel('True efficiency', fontsize=18)\n",
    "plt.xlabel('Predicted efficiency', fontsize=18)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.yticks(fontsize=14)\n",
    "plt.title(' keras regressor ',fontsize=20)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.67821818  0.57858379  0.41760599  0.53988039  0.41102973]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "regressor = LinearRegression()\n",
    "regressor.fit(X_train, y_train.values.ravel())\n",
    "\n",
    "results = cross_val_score(regressor,np.array(X), np.array(y), cv=kfold)\n",
    "print(np.sqrt(results))\n",
    "\n",
    "y_predictions = regressor.predict(np.array(X_test))\n",
    "pred_train = regressor.predict(np.array(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c66d314748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (10.0, 10.0)\n",
    "plt.scatter(pred_train, y_train, c='r', s=40, label='training')\n",
    "plt.scatter(y_predictions, y_test, c='b', s=40, label='testing')\n",
    "plt.plot([0, 60], [-0,60], 'g-', label='true line')\n",
    "plt.legend(loc=4, frameon=True, title='Legend', fontsize=14);\n",
    "plt.ylabel('True efficiency', fontsize=18)\n",
    "plt.xlabel('Predicted efficiency', fontsize=18)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.yticks(fontsize=14)\n",
    "plt.title(' Linear regressor ',fontsize=20)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}