lstm_garch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "import yfinance as yf\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>return</th>\n",
       "      <th>log_return</th>\n",
       "      <th>realized volatility</th>\n",
       "      <th>alpha[1]</th>\n",
       "      <th>beta[1]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1928-03-05</th>\n",
       "      <td>17.570000</td>\n",
       "      <td>17.570000</td>\n",
       "      <td>17.570000</td>\n",
       "      <td>17.570000</td>\n",
       "      <td>17.570000</td>\n",
       "      <td>1.560696</td>\n",
       "      <td>2.234219</td>\n",
       "      <td>0.702718</td>\n",
       "      <td>1.190769e-16</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1928-03-06</th>\n",
       "      <td>17.670000</td>\n",
       "      <td>17.670000</td>\n",
       "      <td>17.670000</td>\n",
       "      <td>17.670000</td>\n",
       "      <td>17.670000</td>\n",
       "      <td>0.569154</td>\n",
       "      <td>0.818788</td>\n",
       "      <td>0.710786</td>\n",
       "      <td>4.727986e-15</td>\n",
       "      <td>0.079529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1928-03-07</th>\n",
       "      <td>17.580000</td>\n",
       "      <td>17.580000</td>\n",
       "      <td>17.580000</td>\n",
       "      <td>17.580000</td>\n",
       "      <td>17.580000</td>\n",
       "      <td>-0.509339</td>\n",
       "      <td>-0.736698</td>\n",
       "      <td>0.709591</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.516325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1928-03-08</th>\n",
       "      <td>17.639999</td>\n",
       "      <td>17.639999</td>\n",
       "      <td>17.639999</td>\n",
       "      <td>17.639999</td>\n",
       "      <td>17.639999</td>\n",
       "      <td>0.341294</td>\n",
       "      <td>0.491545</td>\n",
       "      <td>0.651001</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.912505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1928-03-09</th>\n",
       "      <td>17.930000</td>\n",
       "      <td>17.930000</td>\n",
       "      <td>17.930000</td>\n",
       "      <td>17.930000</td>\n",
       "      <td>17.930000</td>\n",
       "      <td>1.643996</td>\n",
       "      <td>2.352500</td>\n",
       "      <td>0.732310</td>\n",
       "      <td>6.237012e-10</td>\n",
       "      <td>1.000000</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>2022-03-28</th>\n",
       "      <td>4541.089844</td>\n",
       "      <td>4575.649902</td>\n",
       "      <td>4517.689941</td>\n",
       "      <td>4575.520020</td>\n",
       "      <td>4575.520020</td>\n",
       "      <td>0.714496</td>\n",
       "      <td>1.027134</td>\n",
       "      <td>1.482212</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.971109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-29</th>\n",
       "      <td>4602.859863</td>\n",
       "      <td>4637.299805</td>\n",
       "      <td>4589.660156</td>\n",
       "      <td>4631.600098</td>\n",
       "      <td>4631.600098</td>\n",
       "      <td>1.225655</td>\n",
       "      <td>1.757497</td>\n",
       "      <td>1.434337</td>\n",
       "      <td>1.019311e-13</td>\n",
       "      <td>0.971074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-30</th>\n",
       "      <td>4624.200195</td>\n",
       "      <td>4627.770020</td>\n",
       "      <td>4581.319824</td>\n",
       "      <td>4602.450195</td>\n",
       "      <td>4602.450195</td>\n",
       "      <td>-0.629370</td>\n",
       "      <td>-0.910858</td>\n",
       "      <td>1.443096</td>\n",
       "      <td>5.266536e-18</td>\n",
       "      <td>0.962702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-31</th>\n",
       "      <td>4599.020020</td>\n",
       "      <td>4603.069824</td>\n",
       "      <td>4530.410156</td>\n",
       "      <td>4530.410156</td>\n",
       "      <td>4530.410156</td>\n",
       "      <td>-1.565254</td>\n",
       "      <td>-2.276044</td>\n",
       "      <td>1.444158</td>\n",
       "      <td>1.979747e-14</td>\n",
       "      <td>0.974588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-04-01</th>\n",
       "      <td>4540.319824</td>\n",
       "      <td>4548.700195</td>\n",
       "      <td>4507.569824</td>\n",
       "      <td>4545.859863</td>\n",
       "      <td>4545.859863</td>\n",
       "      <td>0.341022</td>\n",
       "      <td>0.491154</td>\n",
       "      <td>1.398410</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.969119</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>23634 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Open         High          Low        Close    Adj Close  \\\n",
       "Date                                                                          \n",
       "1928-03-05    17.570000    17.570000    17.570000    17.570000    17.570000   \n",
       "1928-03-06    17.670000    17.670000    17.670000    17.670000    17.670000   \n",
       "1928-03-07    17.580000    17.580000    17.580000    17.580000    17.580000   \n",
       "1928-03-08    17.639999    17.639999    17.639999    17.639999    17.639999   \n",
       "1928-03-09    17.930000    17.930000    17.930000    17.930000    17.930000   \n",
       "...                 ...          ...          ...          ...          ...   \n",
       "2022-03-28  4541.089844  4575.649902  4517.689941  4575.520020  4575.520020   \n",
       "2022-03-29  4602.859863  4637.299805  4589.660156  4631.600098  4631.600098   \n",
       "2022-03-30  4624.200195  4627.770020  4581.319824  4602.450195  4602.450195   \n",
       "2022-03-31  4599.020020  4603.069824  4530.410156  4530.410156  4530.410156   \n",
       "2022-04-01  4540.319824  4548.700195  4507.569824  4545.859863  4545.859863   \n",
       "\n",
       "              return  log_return  realized volatility      alpha[1]   beta[1]  \n",
       "Date                                                                           \n",
       "1928-03-05  1.560696    2.234219             0.702718  1.190769e-16  1.000000  \n",
       "1928-03-06  0.569154    0.818788             0.710786  4.727986e-15  0.079529  \n",
       "1928-03-07 -0.509339   -0.736698             0.709591  0.000000e+00  0.516325  \n",
       "1928-03-08  0.341294    0.491545             0.651001  0.000000e+00  0.912505  \n",
       "1928-03-09  1.643996    2.352500             0.732310  6.237012e-10  1.000000  \n",
       "...              ...         ...                  ...           ...       ...  \n",
       "2022-03-28  0.714496    1.027134             1.482212  0.000000e+00  0.971109  \n",
       "2022-03-29  1.225655    1.757497             1.434337  1.019311e-13  0.971074  \n",
       "2022-03-30 -0.629370   -0.910858             1.443096  5.266536e-18  0.962702  \n",
       "2022-03-31 -1.565254   -2.276044             1.444158  1.979747e-14  0.974588  \n",
       "2022-04-01  0.341022    0.491154             1.398410  0.000000e+00  0.969119  \n",
       "\n",
       "[23634 rows x 10 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('lstm_garch.csv',index_col='Date')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_dim: the dimension of the input X\n",
    "# time_steps: predict of n days\n",
    "# if x_dim = 22, time steps = 1, it means we use X1 = [p1,p2,...,p22] to predict p23(y1)\n",
    "\n",
    "def create_dataset(data, x_dim, time_steps):\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(data) - x_dim - time_steps):\n",
    "        dataX.append(data[i:(i + x_dim), 0])\n",
    "        dataY.append(data[i + x_dim + time_steps, 0])\n",
    "    return np.array(dataX), np.array(dataY).reshape(-1,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "vol=np.array(df['realized volatility']).reshape(-1,1)\n",
    "alpha = np.array(df['alpha[1]']).reshape(-1,1)\n",
    "beta = np.array(df['beta[1]']).reshape(-1,1)\n",
    "\n",
    "scaler_vol = MinMaxScaler(feature_range=(0,1))\n",
    "scaler_alpha = MinMaxScaler(feature_range=(0,1))\n",
    "scaler_beta = MinMaxScaler(feature_range=(0,1))\n",
    "\n",
    "vol=scaler_vol.fit_transform(vol)\n",
    "alpha=scaler_alpha.fit_transform(alpha)\n",
    "beta=scaler_beta.fit_transform(beta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_dim = 22\n",
    "time_steps = 1\n",
    "x_vol,y = create_dataset(vol,x_dim,time_steps)\n",
    "x_alpha = create_dataset(alpha,x_dim,time_steps)[0]\n",
    "x_beta = create_dataset(beta,x_dim,time_steps)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = np.hstack((x_vol,x_alpha,x_beta,y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((18889, 67), (4722, 67))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratio = 0.8\n",
    "Traindata = data[0:round(len(data)*ratio)]\n",
    "Testdata = data[round(len(data)*ratio):]\n",
    "Traindata.shape, Testdata.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data size: 18889 Train size: 17000 Test size: 1889\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "train, test = train_test_split(Traindata, train_size=0.9)\n",
    "X_train, y_train = train[:,:-1], train[:,-1]\n",
    "X_test, y_test = test[:,:-1], test[:,-1]\n",
    "print('Data size: {} Train size: {} Test size: {}'.format(Traindata.shape[0], X_train.shape[0], X_test.shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------\n",
      "X_train shape: (17000, 66, 1)\n",
      "------------------------------------------------------\n",
      "X_test shape: (1889, 66, 1)\n"
     ]
    }
   ],
   "source": [
    "X_train = X_train.reshape((X_train.shape[0],X_train.shape[1],1))\n",
    "X_test = X_test.reshape((X_test.shape[0],X_train.shape[1],1))\n",
    "print('------------------------------------------------------')\n",
    "print('X_train shape: {}'.format(X_train.shape))\n",
    "print('------------------------------------------------------')\n",
    "print('X_test shape: {}'.format(X_test.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " lstm_1 (LSTM)               (None, 32)                4352      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 4,385\n",
      "Trainable params: 4,385\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(32, input_shape=(x_dim*3, 1), return_sequences=False))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mean_squared_error', optimizer='adam',metrics=['mae', 'mape','msle'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "207/207 [==============================] - 3s 11ms/step - loss: 0.0123 - mae: 0.0731 - mape: 82.3256 - msle: 0.0078 - val_loss: 0.0118 - val_mae: 0.0691 - val_mape: 31003.7520 - val_msle: 0.0075\n",
      "Epoch 2/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0112 - mae: 0.0696 - mape: 79.7331 - msle: 0.0071 - val_loss: 0.0060 - val_mae: 0.0510 - val_mape: 28876.0430 - val_msle: 0.0037\n",
      "Epoch 3/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0044 - mae: 0.0444 - mape: 50.7959 - msle: 0.0028 - val_loss: 0.0039 - val_mae: 0.0431 - val_mape: 27462.3828 - val_msle: 0.0025\n",
      "Epoch 4/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0038 - mae: 0.0408 - mape: 45.4559 - msle: 0.0024 - val_loss: 0.0035 - val_mae: 0.0414 - val_mape: 25130.6387 - val_msle: 0.0023\n",
      "Epoch 5/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0035 - mae: 0.0393 - mape: 42.7667 - msle: 0.0022 - val_loss: 0.0033 - val_mae: 0.0400 - val_mape: 25200.2930 - val_msle: 0.0021\n",
      "Epoch 6/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0034 - mae: 0.0382 - mape: 41.0749 - msle: 0.0021 - val_loss: 0.0032 - val_mae: 0.0371 - val_mape: 23187.5684 - val_msle: 0.0020\n",
      "Epoch 7/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0032 - mae: 0.0377 - mape: 40.4830 - msle: 0.0021 - val_loss: 0.0030 - val_mae: 0.0377 - val_mape: 22342.0234 - val_msle: 0.0019\n",
      "Epoch 8/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0031 - mae: 0.0367 - mape: 39.2180 - msle: 0.0020 - val_loss: 0.0029 - val_mae: 0.0356 - val_mape: 21469.5859 - val_msle: 0.0019\n",
      "Epoch 9/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 0.0031 - mae: 0.0367 - mape: 39.1888 - msle: 0.0020 - val_loss: 0.0028 - val_mae: 0.0342 - val_mape: 19581.3594 - val_msle: 0.0017\n",
      "Epoch 10/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0027 - mae: 0.0339 - mape: 35.9790 - msle: 0.0017 - val_loss: 0.0023 - val_mae: 0.0325 - val_mape: 10936.3105 - val_msle: 0.0015\n",
      "Epoch 11/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0022 - mae: 0.0310 - mape: 31.8641 - msle: 0.0014 - val_loss: 0.0017 - val_mae: 0.0277 - val_mape: 8224.9551 - val_msle: 0.0011\n",
      "Epoch 12/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 0.0020 - mae: 0.0292 - mape: 29.5014 - msle: 0.0013 - val_loss: 0.0016 - val_mae: 0.0268 - val_mape: 11302.1338 - val_msle: 0.0011\n",
      "Epoch 13/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0016 - mae: 0.0256 - mape: 25.4436 - msle: 0.0010 - val_loss: 0.0019 - val_mae: 0.0262 - val_mape: 10444.5176 - val_msle: 0.0011\n",
      "Epoch 14/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 0.0015 - mae: 0.0246 - mape: 24.2981 - msle: 9.5024e-04 - val_loss: 0.0013 - val_mae: 0.0231 - val_mape: 11365.6680 - val_msle: 8.4662e-04\n",
      "Epoch 15/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0014 - mae: 0.0235 - mape: 23.2631 - msle: 8.7503e-04 - val_loss: 0.0015 - val_mae: 0.0260 - val_mape: 11332.6299 - val_msle: 9.6599e-04\n",
      "Epoch 16/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0015 - mae: 0.0246 - mape: 24.5395 - msle: 9.4792e-04 - val_loss: 0.0013 - val_mae: 0.0233 - val_mape: 13860.4150 - val_msle: 8.3566e-04\n",
      "Epoch 17/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0013 - mae: 0.0233 - mape: 23.1668 - msle: 8.5266e-04 - val_loss: 0.0015 - val_mae: 0.0252 - val_mape: 9889.4561 - val_msle: 9.6889e-04\n",
      "Epoch 18/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0014 - mae: 0.0234 - mape: 22.8434 - msle: 8.8678e-04 - val_loss: 0.0015 - val_mae: 0.0252 - val_mape: 13745.0283 - val_msle: 9.6230e-04\n",
      "Epoch 19/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0014 - mae: 0.0235 - mape: 23.3806 - msle: 8.7001e-04 - val_loss: 0.0016 - val_mae: 0.0247 - val_mape: 11670.7637 - val_msle: 9.9647e-04\n",
      "Epoch 20/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0013 - mae: 0.0234 - mape: 23.3093 - msle: 8.4624e-04 - val_loss: 0.0019 - val_mae: 0.0294 - val_mape: 9724.2295 - val_msle: 0.0012\n",
      "Epoch 21/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0013 - mae: 0.0232 - mape: 23.2644 - msle: 8.3706e-04 - val_loss: 0.0013 - val_mae: 0.0247 - val_mape: 9187.4629 - val_msle: 8.7022e-04\n",
      "Epoch 22/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0012 - mae: 0.0220 - mape: 22.2412 - msle: 7.6785e-04 - val_loss: 0.0012 - val_mae: 0.0240 - val_mape: 11261.7861 - val_msle: 8.0196e-04\n",
      "Epoch 23/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0012 - mae: 0.0220 - mape: 22.0140 - msle: 7.6290e-04 - val_loss: 0.0011 - val_mae: 0.0213 - val_mape: 9097.9873 - val_msle: 7.2096e-04\n",
      "Epoch 24/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0206 - mape: 20.1808 - msle: 6.9213e-04 - val_loss: 0.0011 - val_mae: 0.0218 - val_mape: 10096.6465 - val_msle: 7.4332e-04\n",
      "Epoch 25/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0204 - mape: 19.8706 - msle: 6.7704e-04 - val_loss: 0.0014 - val_mae: 0.0289 - val_mape: 13103.1875 - val_msle: 9.8845e-04\n",
      "Epoch 26/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0216 - mape: 21.6458 - msle: 7.3074e-04 - val_loss: 0.0012 - val_mae: 0.0240 - val_mape: 13995.2881 - val_msle: 7.8827e-04\n",
      "Epoch 27/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0212 - mape: 20.9337 - msle: 7.1106e-04 - val_loss: 9.6995e-04 - val_mae: 0.0193 - val_mape: 7723.6470 - val_msle: 6.4159e-04\n",
      "Epoch 28/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0206 - mape: 20.4084 - msle: 6.8251e-04 - val_loss: 0.0010 - val_mae: 0.0212 - val_mape: 9216.3271 - val_msle: 7.0715e-04\n",
      "Epoch 29/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 0.0011 - mae: 0.0212 - mape: 20.9335 - msle: 7.0605e-04 - val_loss: 0.0013 - val_mae: 0.0221 - val_mape: 10067.8418 - val_msle: 7.9459e-04\n",
      "Epoch 30/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0210 - mape: 20.8012 - msle: 7.0710e-04 - val_loss: 0.0010 - val_mae: 0.0199 - val_mape: 8485.8057 - val_msle: 6.5569e-04\n",
      "Epoch 31/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.8937e-04 - mae: 0.0198 - mape: 19.5409 - msle: 6.3596e-04 - val_loss: 0.0010 - val_mae: 0.0225 - val_mape: 12768.6953 - val_msle: 6.9487e-04\n",
      "Epoch 32/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 9.9646e-04 - mae: 0.0200 - mape: 19.8639 - msle: 6.4520e-04 - val_loss: 9.2384e-04 - val_mae: 0.0186 - val_mape: 9185.0010 - val_msle: 6.1023e-04\n",
      "Epoch 33/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.9682e-04 - mae: 0.0200 - mape: 19.8942 - msle: 6.4395e-04 - val_loss: 0.0010 - val_mae: 0.0220 - val_mape: 11472.3115 - val_msle: 6.8420e-04\n",
      "Epoch 34/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0010 - mae: 0.0203 - mape: 20.2359 - msle: 6.6208e-04 - val_loss: 9.3371e-04 - val_mae: 0.0210 - val_mape: 11878.8193 - val_msle: 6.3695e-04\n",
      "Epoch 35/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0011 - mae: 0.0211 - mape: 21.3227 - msle: 6.9109e-04 - val_loss: 8.9799e-04 - val_mae: 0.0191 - val_mape: 9316.6660 - val_msle: 5.9223e-04\n",
      "Epoch 36/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.8659e-04 - mae: 0.0199 - mape: 19.7188 - msle: 6.3262e-04 - val_loss: 8.9181e-04 - val_mae: 0.0184 - val_mape: 11079.7129 - val_msle: 5.8003e-04\n",
      "Epoch 37/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.7859e-04 - mae: 0.0197 - mape: 19.5181 - msle: 6.2638e-04 - val_loss: 8.4826e-04 - val_mae: 0.0183 - val_mape: 11097.0332 - val_msle: 5.6384e-04\n",
      "Epoch 38/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.6858e-04 - mae: 0.0197 - mape: 19.6753 - msle: 6.2627e-04 - val_loss: 8.9746e-04 - val_mae: 0.0189 - val_mape: 10586.6631 - val_msle: 5.9757e-04\n",
      "Epoch 39/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 0.0010 - mae: 0.0202 - mape: 20.4075 - msle: 6.5379e-04 - val_loss: 9.7549e-04 - val_mae: 0.0199 - val_mape: 8745.3438 - val_msle: 6.3861e-04\n",
      "Epoch 40/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.2943e-04 - mae: 0.0191 - mape: 18.8160 - msle: 5.9811e-04 - val_loss: 9.0041e-04 - val_mae: 0.0186 - val_mape: 10283.3281 - val_msle: 5.9043e-04\n",
      "Epoch 41/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 0.0010 - mae: 0.0203 - mape: 20.3210 - msle: 6.4978e-04 - val_loss: 0.0011 - val_mae: 0.0231 - val_mape: 9807.8789 - val_msle: 7.0089e-04\n",
      "Epoch 42/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.9067e-04 - mae: 0.0200 - mape: 20.0139 - msle: 6.3818e-04 - val_loss: 8.2913e-04 - val_mae: 0.0179 - val_mape: 9512.0771 - val_msle: 5.4692e-04\n",
      "Epoch 43/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.3647e-04 - mae: 0.0192 - mape: 19.1023 - msle: 6.0064e-04 - val_loss: 8.7359e-04 - val_mae: 0.0198 - val_mape: 8144.9287 - val_msle: 5.8600e-04\n",
      "Epoch 44/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.2595e-04 - mae: 0.0194 - mape: 19.6557 - msle: 6.0188e-04 - val_loss: 9.1661e-04 - val_mae: 0.0188 - val_mape: 5515.5708 - val_msle: 5.9412e-04\n",
      "Epoch 45/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.1323e-04 - mae: 0.0190 - mape: 18.9303 - msle: 5.8717e-04 - val_loss: 8.0853e-04 - val_mae: 0.0176 - val_mape: 8526.1660 - val_msle: 5.3133e-04\n",
      "Epoch 46/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 9.4999e-04 - mae: 0.0194 - mape: 19.3757 - msle: 6.0645e-04 - val_loss: 8.1686e-04 - val_mae: 0.0177 - val_mape: 8892.2217 - val_msle: 5.4191e-04\n",
      "Epoch 47/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.8545e-04 - mae: 0.0186 - mape: 18.5717 - msle: 5.7018e-04 - val_loss: 8.1141e-04 - val_mae: 0.0175 - val_mape: 10988.5156 - val_msle: 5.3376e-04\n",
      "Epoch 48/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.6593e-04 - mae: 0.0186 - mape: 18.7246 - msle: 5.6123e-04 - val_loss: 8.3745e-04 - val_mae: 0.0179 - val_mape: 6492.8848 - val_msle: 5.4280e-04\n",
      "Epoch 49/100\n",
      "207/207 [==============================] - 2s 11ms/step - loss: 8.5774e-04 - mae: 0.0182 - mape: 18.0875 - msle: 5.4933e-04 - val_loss: 8.3628e-04 - val_mae: 0.0184 - val_mape: 5461.4102 - val_msle: 5.5637e-04\n",
      "Epoch 50/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.7938e-04 - mae: 0.0189 - mape: 19.0536 - msle: 5.7207e-04 - val_loss: 8.6406e-04 - val_mae: 0.0191 - val_mape: 9131.2959 - val_msle: 5.8371e-04\n",
      "Epoch 51/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.6315e-04 - mae: 0.0184 - mape: 18.5162 - msle: 5.5441e-04 - val_loss: 8.5749e-04 - val_mae: 0.0184 - val_mape: 10210.0029 - val_msle: 5.6284e-04\n",
      "Epoch 52/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.5123e-04 - mae: 0.0181 - mape: 18.0303 - msle: 5.4428e-04 - val_loss: 7.4421e-04 - val_mae: 0.0171 - val_mape: 8370.0635 - val_msle: 4.8808e-04\n",
      "Epoch 53/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.6594e-04 - mae: 0.0185 - mape: 18.3782 - msle: 5.5413e-04 - val_loss: 7.6430e-04 - val_mae: 0.0178 - val_mape: 6065.4209 - val_msle: 5.1630e-04\n",
      "Epoch 54/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 9.1879e-04 - mae: 0.0188 - mape: 18.6809 - msle: 5.8128e-04 - val_loss: 0.0012 - val_mae: 0.0236 - val_mape: 11709.7598 - val_msle: 7.4996e-04\n",
      "Epoch 55/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.4619e-04 - mae: 0.0182 - mape: 18.1485 - msle: 5.4223e-04 - val_loss: 8.4425e-04 - val_mae: 0.0193 - val_mape: 5410.2354 - val_msle: 5.7624e-04\n",
      "Epoch 56/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.4591e-04 - mae: 0.0184 - mape: 18.3684 - msle: 5.4387e-04 - val_loss: 8.0795e-04 - val_mae: 0.0178 - val_mape: 9411.0684 - val_msle: 5.2571e-04\n",
      "Epoch 57/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.4781e-04 - mae: 0.0184 - mape: 18.4652 - msle: 5.4560e-04 - val_loss: 8.4931e-04 - val_mae: 0.0181 - val_mape: 7483.0928 - val_msle: 5.2711e-04\n",
      "Epoch 58/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.3608e-04 - mae: 0.0181 - mape: 18.1965 - msle: 5.3798e-04 - val_loss: 8.5560e-04 - val_mae: 0.0184 - val_mape: 8144.4146 - val_msle: 5.5725e-04\n",
      "Epoch 59/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.4007e-04 - mae: 0.0183 - mape: 18.2886 - msle: 5.3761e-04 - val_loss: 8.8134e-04 - val_mae: 0.0193 - val_mape: 12597.6973 - val_msle: 5.7409e-04\n",
      "Epoch 60/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.4188e-04 - mae: 0.0183 - mape: 18.3086 - msle: 5.3877e-04 - val_loss: 8.6366e-04 - val_mae: 0.0198 - val_mape: 7978.2378 - val_msle: 5.7616e-04\n",
      "Epoch 61/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.9148e-04 - mae: 0.0176 - mape: 17.5344 - msle: 5.0661e-04 - val_loss: 7.0387e-04 - val_mae: 0.0168 - val_mape: 8296.7402 - val_msle: 4.7000e-04\n",
      "Epoch 62/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.0237e-04 - mae: 0.0178 - mape: 17.9584 - msle: 5.1478e-04 - val_loss: 7.9633e-04 - val_mae: 0.0198 - val_mape: 6008.1113 - val_msle: 5.4077e-04\n",
      "Epoch 63/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.7705e-04 - mae: 0.0174 - mape: 17.4222 - msle: 4.9870e-04 - val_loss: 6.8677e-04 - val_mae: 0.0165 - val_mape: 10522.8467 - val_msle: 4.6124e-04\n",
      "Epoch 64/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.7515e-04 - mae: 0.0174 - mape: 17.3332 - msle: 4.9731e-04 - val_loss: 8.0229e-04 - val_mae: 0.0178 - val_mape: 10608.0791 - val_msle: 5.3361e-04\n",
      "Epoch 65/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 8.1078e-04 - mae: 0.0179 - mape: 18.0119 - msle: 5.1799e-04 - val_loss: 7.7273e-04 - val_mae: 0.0183 - val_mape: 8155.3457 - val_msle: 5.2857e-04\n",
      "Epoch 66/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.6427e-04 - mae: 0.0173 - mape: 17.4843 - msle: 4.9226e-04 - val_loss: 7.3284e-04 - val_mae: 0.0164 - val_mape: 6833.4556 - val_msle: 4.6986e-04\n",
      "Epoch 67/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.3532e-04 - mae: 0.0167 - mape: 16.4354 - msle: 4.6793e-04 - val_loss: 8.2211e-04 - val_mae: 0.0206 - val_mape: 10353.1807 - val_msle: 5.5853e-04\n",
      "Epoch 68/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.4116e-04 - mae: 0.0172 - mape: 17.1941 - msle: 4.7981e-04 - val_loss: 8.3907e-04 - val_mae: 0.0207 - val_mape: 7658.0083 - val_msle: 5.6535e-04\n",
      "Epoch 69/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.1323e-04 - mae: 0.0168 - mape: 17.0009 - msle: 4.6138e-04 - val_loss: 9.5180e-04 - val_mae: 0.0187 - val_mape: 2926.2454 - val_msle: 5.9256e-04\n",
      "Epoch 70/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.4898e-04 - mae: 0.0171 - mape: 17.0301 - msle: 4.8061e-04 - val_loss: 6.7963e-04 - val_mae: 0.0177 - val_mape: 8970.4570 - val_msle: 4.5929e-04\n",
      "Epoch 71/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.3924e-04 - mae: 0.0169 - mape: 17.1611 - msle: 4.7283e-04 - val_loss: 6.8098e-04 - val_mae: 0.0161 - val_mape: 5649.8486 - val_msle: 4.4957e-04\n",
      "Epoch 72/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.3079e-04 - mae: 0.0170 - mape: 17.0063 - msle: 4.6800e-04 - val_loss: 6.5498e-04 - val_mae: 0.0174 - val_mape: 9373.4512 - val_msle: 4.4650e-04\n",
      "Epoch 73/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.1004e-04 - mae: 0.0166 - mape: 16.5768 - msle: 4.5498e-04 - val_loss: 7.2252e-04 - val_mae: 0.0166 - val_mape: 6959.6230 - val_msle: 4.6917e-04\n",
      "Epoch 74/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.6514e-04 - mae: 0.0161 - mape: 16.1042 - msle: 4.2830e-04 - val_loss: 7.3448e-04 - val_mae: 0.0174 - val_mape: 4567.4292 - val_msle: 4.9072e-04\n",
      "Epoch 75/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.1121e-04 - mae: 0.0168 - mape: 16.9536 - msle: 4.5723e-04 - val_loss: 8.6905e-04 - val_mae: 0.0203 - val_mape: 7600.0854 - val_msle: 5.9964e-04\n",
      "Epoch 76/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.9434e-04 - mae: 0.0166 - mape: 16.6744 - msle: 4.4563e-04 - val_loss: 7.0969e-04 - val_mae: 0.0159 - val_mape: 5736.1533 - val_msle: 4.5074e-04\n",
      "Epoch 77/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.8373e-04 - mae: 0.0163 - mape: 16.3492 - msle: 4.3990e-04 - val_loss: 9.8237e-04 - val_mae: 0.0224 - val_mape: 5894.5396 - val_msle: 6.8980e-04\n",
      "Epoch 78/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 7.0596e-04 - mae: 0.0169 - mape: 17.1587 - msle: 4.5601e-04 - val_loss: 6.0732e-04 - val_mae: 0.0153 - val_mape: 7321.7310 - val_msle: 4.0698e-04\n",
      "Epoch 79/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.3136e-04 - mae: 0.0154 - mape: 15.3161 - msle: 4.0516e-04 - val_loss: 5.9040e-04 - val_mae: 0.0147 - val_mape: 7077.8979 - val_msle: 3.8923e-04\n",
      "Epoch 80/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.6353e-04 - mae: 0.0159 - mape: 15.8494 - msle: 4.2618e-04 - val_loss: 6.0209e-04 - val_mae: 0.0154 - val_mape: 9462.8623 - val_msle: 4.0511e-04\n",
      "Epoch 81/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.5008e-04 - mae: 0.0160 - mape: 15.9784 - msle: 4.1938e-04 - val_loss: 6.0950e-04 - val_mae: 0.0152 - val_mape: 7691.4717 - val_msle: 4.0795e-04\n",
      "Epoch 82/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.4865e-04 - mae: 0.0158 - mape: 15.5040 - msle: 4.1375e-04 - val_loss: 7.7138e-04 - val_mae: 0.0196 - val_mape: 5714.5205 - val_msle: 5.4080e-04\n",
      "Epoch 83/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.7509e-04 - mae: 0.0163 - mape: 16.1364 - msle: 4.3351e-04 - val_loss: 6.8200e-04 - val_mae: 0.0176 - val_mape: 7535.5532 - val_msle: 4.7151e-04\n",
      "Epoch 84/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.5900e-04 - mae: 0.0160 - mape: 16.0814 - msle: 4.2370e-04 - val_loss: 6.2329e-04 - val_mae: 0.0159 - val_mape: 6810.5850 - val_msle: 4.2075e-04\n",
      "Epoch 85/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.5336e-04 - mae: 0.0160 - mape: 16.2129 - msle: 4.2136e-04 - val_loss: 5.9587e-04 - val_mae: 0.0157 - val_mape: 6210.4844 - val_msle: 3.9605e-04\n",
      "Epoch 86/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.0295e-04 - mae: 0.0150 - mape: 14.7404 - msle: 3.8395e-04 - val_loss: 5.6753e-04 - val_mae: 0.0146 - val_mape: 8576.6338 - val_msle: 3.7669e-04\n",
      "Epoch 87/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.5730e-04 - mae: 0.0163 - mape: 16.4918 - msle: 4.2603e-04 - val_loss: 6.1973e-04 - val_mae: 0.0172 - val_mape: 11722.5352 - val_msle: 4.2480e-04\n",
      "Epoch 88/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.4057e-04 - mae: 0.0160 - mape: 16.0378 - msle: 4.1346e-04 - val_loss: 8.2860e-04 - val_mae: 0.0191 - val_mape: 4052.3704 - val_msle: 5.4890e-04\n",
      "Epoch 89/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.9442e-04 - mae: 0.0148 - mape: 14.6396 - msle: 3.7935e-04 - val_loss: 8.1099e-04 - val_mae: 0.0183 - val_mape: 4069.5129 - val_msle: 5.3023e-04\n",
      "Epoch 90/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.4333e-04 - mae: 0.0159 - mape: 15.8166 - msle: 4.1359e-04 - val_loss: 6.5078e-04 - val_mae: 0.0172 - val_mape: 9215.1094 - val_msle: 4.4940e-04\n",
      "Epoch 91/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.0408e-04 - mae: 0.0155 - mape: 15.5915 - msle: 3.9170e-04 - val_loss: 6.0928e-04 - val_mae: 0.0156 - val_mape: 10516.4111 - val_msle: 4.0139e-04\n",
      "Epoch 92/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.0557e-04 - mae: 0.0152 - mape: 14.9178 - msle: 3.8747e-04 - val_loss: 5.7723e-04 - val_mae: 0.0148 - val_mape: 7900.8242 - val_msle: 3.7564e-04\n",
      "Epoch 93/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.2070e-04 - mae: 0.0154 - mape: 15.2638 - msle: 3.9454e-04 - val_loss: 7.9467e-04 - val_mae: 0.0192 - val_mape: 7710.9458 - val_msle: 5.2977e-04\n",
      "Epoch 94/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.7590e-04 - mae: 0.0147 - mape: 14.5549 - msle: 3.6968e-04 - val_loss: 6.7205e-04 - val_mae: 0.0166 - val_mape: 10708.5635 - val_msle: 4.4796e-04\n",
      "Epoch 95/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 6.3577e-04 - mae: 0.0156 - mape: 15.3419 - msle: 4.0461e-04 - val_loss: 5.2461e-04 - val_mae: 0.0138 - val_mape: 7254.6841 - val_msle: 3.4668e-04\n",
      "Epoch 96/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.7920e-04 - mae: 0.0148 - mape: 14.4698 - msle: 3.6948e-04 - val_loss: 6.2582e-04 - val_mae: 0.0151 - val_mape: 8715.9502 - val_msle: 3.9535e-04\n",
      "Epoch 97/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.8169e-04 - mae: 0.0150 - mape: 14.8277 - msle: 3.7320e-04 - val_loss: 5.2327e-04 - val_mae: 0.0151 - val_mape: 8020.8809 - val_msle: 3.5184e-04\n",
      "Epoch 98/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.6023e-04 - mae: 0.0146 - mape: 14.3560 - msle: 3.5899e-04 - val_loss: 5.2590e-04 - val_mae: 0.0138 - val_mape: 8417.0996 - val_msle: 3.4767e-04\n",
      "Epoch 99/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.5581e-04 - mae: 0.0145 - mape: 14.2172 - msle: 3.5582e-04 - val_loss: 6.1823e-04 - val_mae: 0.0158 - val_mape: 9284.2637 - val_msle: 4.0014e-04\n",
      "Epoch 100/100\n",
      "207/207 [==============================] - 2s 10ms/step - loss: 5.4286e-04 - mae: 0.0141 - mape: 13.8372 - msle: 3.4684e-04 - val_loss: 5.4669e-04 - val_mae: 0.0142 - val_mape: 5103.2637 - val_msle: 3.6109e-04\n",
      "266/266 [==============================] - 1s 3ms/step - loss: 5.4569e-04 - mae: 0.0137 - mape: 1143.3671 - msle: 3.4871e-04\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, epochs=100, batch_size=64, verbose=1, validation_split=2/9)\n",
    "score = model.evaluate(X_train, y_train, batch_size=64, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x450 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x450 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x450 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x450 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def visualize_loss(history, evaluation):\n",
    "    train = history.history[evaluation]\n",
    "    val = history.history['val_'+evaluation]\n",
    "    epochs = range(len(train))\n",
    "    plt.figure(figsize=(8,5),dpi=90)\n",
    "    plt.plot(epochs, train, label=evaluation)\n",
    "    plt.plot(epochs, val, label='val_'+evaluation)\n",
    "    plt.title(evaluation)\n",
    "    plt.grid()\n",
    "    plt.xlabel(\"Epochs\")\n",
    "    plt.ylabel(evaluation)\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "for eva in ['loss','mae', 'mape','msle']:\n",
    "    visualize_loss(history, eva)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "Test_X = Testdata[:,:-1]\n",
    "Test_y = Testdata[:,-1]\n",
    "Test_y_pre = model.predict(Test_X)\n",
    "Test_y_pre = scaler_vol.inverse_transform(Test_y_pre)\n",
    "Test_y = scaler_vol.inverse_transform(Test_y.reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2: 95.2143%\n",
      "mean squared error: 2.3961%\n",
      "mean absolute error: 9.0086%\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "mse = metrics.mean_squared_error(Test_y, Test_y_pre)\n",
    "mae = metrics.mean_absolute_error(Test_y, Test_y_pre)\n",
    "r2 = metrics.r2_score(Test_y, Test_y_pre)\n",
    "print('r2: {:.4f}%'.format(r2*100))\n",
    "print('mean squared error: {:.4f}%'.format(mse*100))\n",
    "print('mean absolute error: {:.4f}%'.format(mae*100))"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "8b027f683e19d78b27f3a066295e9d20c15df7044c54ff64fda32a2fbf691de1"
  },
  "kernelspec": {
   "display_name": "Python 3.9.12 ('tensorflow')",
   "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": "3.9.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}