LSTM-Attention-feature.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#导入必要的库\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn import preprocessing\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from math import sqrt\n",
    "from tensorflow.keras.layers import *\n",
    "from tensorflow.keras.models import *\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "import tensorflow.keras.backend as K\n",
    "plt.style.use('seaborn')\n",
    "np.random.seed(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "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>station_id</th>\n",
       "      <th>time</th>\n",
       "      <th>PM25_Concentration</th>\n",
       "      <th>PM10_Concentration</th>\n",
       "      <th>NO2_Concentration</th>\n",
       "      <th>CO_Concentration</th>\n",
       "      <th>O3_Concentration</th>\n",
       "      <th>SO2_Concentration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>164466</th>\n",
       "      <td>1036</td>\n",
       "      <td>2015/4/30 16:00</td>\n",
       "      <td>82</td>\n",
       "      <td>123.0</td>\n",
       "      <td>12.1</td>\n",
       "      <td>0.5</td>\n",
       "      <td>187.5</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164467</th>\n",
       "      <td>1036</td>\n",
       "      <td>2015/4/30 17:00</td>\n",
       "      <td>81</td>\n",
       "      <td>87.1</td>\n",
       "      <td>11.1</td>\n",
       "      <td>0.5</td>\n",
       "      <td>192.3</td>\n",
       "      <td>16.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164468</th>\n",
       "      <td>1036</td>\n",
       "      <td>2015/4/30 18:00</td>\n",
       "      <td>79</td>\n",
       "      <td>107.6</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>192.1</td>\n",
       "      <td>18.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164469</th>\n",
       "      <td>1036</td>\n",
       "      <td>2015/4/30 19:00</td>\n",
       "      <td>77</td>\n",
       "      <td>104.6</td>\n",
       "      <td>15.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>180.7</td>\n",
       "      <td>18.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164470</th>\n",
       "      <td>1036</td>\n",
       "      <td>2015/4/30 20:00</td>\n",
       "      <td>94</td>\n",
       "      <td>141.2</td>\n",
       "      <td>26.1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>146.1</td>\n",
       "      <td>16.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        station_id             time  PM25_Concentration  PM10_Concentration  \\\n",
       "164466        1036  2015/4/30 16:00                  82               123.0   \n",
       "164467        1036  2015/4/30 17:00                  81                87.1   \n",
       "164468        1036  2015/4/30 18:00                  79               107.6   \n",
       "164469        1036  2015/4/30 19:00                  77               104.6   \n",
       "164470        1036  2015/4/30 20:00                  94               141.2   \n",
       "\n",
       "        NO2_Concentration  CO_Concentration  O3_Concentration  \\\n",
       "164466               12.1               0.5             187.5   \n",
       "164467               11.1               0.5             192.3   \n",
       "164468               12.0               0.5             192.1   \n",
       "164469               15.5               0.5             180.7   \n",
       "164470               26.1               0.6             146.1   \n",
       "\n",
       "        SO2_Concentration  \n",
       "164466               14.0  \n",
       "164467               16.6  \n",
       "164468               18.2  \n",
       "164469               18.3  \n",
       "164470               16.9  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#设置神经网络参数\n",
    "station_id=1035#选择的监测站\n",
    "lstm_units=16#lstm神经元个数\n",
    "batch_size=64#批训练大小\n",
    "epoch=100#迭代次数\n",
    "test_ratio=.1#测试集比例\n",
    "windows=5#时间窗\n",
    "scale=1.0#归一化参数\n",
    "save_model=('lstm-att-%d.h5'%station_id)#保存下该模型，便于下次加载\n",
    "#读取数据\n",
    "df=pd.read_csv('bjpm25.csv') \n",
    "df=df.iloc[:,:]\n",
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>PM25_Concentration</th>\n",
       "      <th>PM10_Concentration</th>\n",
       "      <th>NO2_Concentration</th>\n",
       "      <th>CO_Concentration</th>\n",
       "      <th>O3_Concentration</th>\n",
       "      <th>SO2_Concentration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>160762</th>\n",
       "      <td>78</td>\n",
       "      <td>97.8</td>\n",
       "      <td>64.4</td>\n",
       "      <td>0.8</td>\n",
       "      <td>173.6</td>\n",
       "      <td>11.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160763</th>\n",
       "      <td>64</td>\n",
       "      <td>96.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>179.0</td>\n",
       "      <td>10.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160764</th>\n",
       "      <td>69</td>\n",
       "      <td>140.2</td>\n",
       "      <td>28.2</td>\n",
       "      <td>0.7</td>\n",
       "      <td>171.8</td>\n",
       "      <td>8.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160765</th>\n",
       "      <td>78</td>\n",
       "      <td>138.3</td>\n",
       "      <td>30.4</td>\n",
       "      <td>0.8</td>\n",
       "      <td>124.1</td>\n",
       "      <td>5.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160766</th>\n",
       "      <td>94</td>\n",
       "      <td>128.4</td>\n",
       "      <td>24.9</td>\n",
       "      <td>1.0</td>\n",
       "      <td>148.6</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        PM25_Concentration  PM10_Concentration  NO2_Concentration  \\\n",
       "160762                  78                97.8               64.4   \n",
       "160763                  64                96.0               45.0   \n",
       "160764                  69               140.2               28.2   \n",
       "160765                  78               138.3               30.4   \n",
       "160766                  94               128.4               24.9   \n",
       "\n",
       "        CO_Concentration  O3_Concentration  SO2_Concentration  \n",
       "160762               0.8             173.6               11.7  \n",
       "160763               0.7             179.0               10.3  \n",
       "160764               0.7             171.8                8.5  \n",
       "160765               0.8             124.1                5.8  \n",
       "160766               1.0             148.6                9.1  "
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#筛选出当前station下的数据\n",
    "df=df[df['station_id']==station_id]\n",
    "df=df.iloc[:,2:]\n",
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3654, 5, 6)\n",
      "(407, 5, 6)\n",
      "(3654,)\n",
      "(407,)\n"
     ]
    }
   ],
   "source": [
    "#把数据处理成lstm接受的输入形式\n",
    "y=df['PM25_Concentration']\n",
    "data = np.array(df)/scale\n",
    "cut = round(test_ratio* data.shape[0])\n",
    "amount_of_features=data.shape[1]\n",
    "lstm_input=[]\n",
    "data_temp=data\n",
    "for i in range(len(data_temp)-windows):\n",
    "    lstm_input.append(data_temp[i:i+windows,:])\n",
    "lstm_input=np.array(lstm_input)\n",
    "#print(lstm_input.shape)\n",
    "lstm_output=y[:-windows]\n",
    "lstm_output=np.array(lstm_output)\n",
    "#print(lstm_output.shape)\n",
    "x_train,y_train,x_test,y_test=\\\n",
    "lstm_input[:-cut,:,:],lstm_output[:-cut:],lstm_input[-cut:,:,:],lstm_output[-cut:]\n",
    "print(x_train.shape)\n",
    "print(x_test.shape)\n",
    "print(y_train.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_23\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_6 (InputLayer)            [(None, 5, 6)]       0                                            \n",
      "__________________________________________________________________________________________________\n",
      "permute_10 (Permute)            (None, 6, 5)         0           input_6[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "lstm_5 (LSTM)                   (None, 6, 16)        1408        permute_10[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "permute_11 (Permute)            (None, 16, 6)        0           lstm_5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "dim_reduction (Lambda)          (None, 6)            0           permute_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "attention_vec (Dense)           (None, 6)            42          dim_reduction[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "attention_weight (Activation)   (None, 6)            0           attention_vec[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "multiply_5 (Multiply)           (None, 6)            0           dim_reduction[0][0]              \n",
      "                                                                 attention_weight[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "dense_5 (Dense)                 (None, 1)            7           multiply_5[0][0]                 \n",
      "==================================================================================================\n",
      "Total params: 1,457\n",
      "Trainable params: 1,457\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#建立模型\n",
    "inputs=Input(shape=(windows,amount_of_features))\n",
    "lstm_inputs=Permute([2,1])(inputs)\n",
    "lstm=LSTM(lstm_units,activation='relu', return_sequences=True)(lstm_inputs)\n",
    "lstm=Permute([2,1])(lstm)#置换维度\n",
    "lstm=Lambda(lambda x: K.mean(x, axis=1), name='dim_reduction')(lstm)\n",
    "attention=Dense(amount_of_features, activation='sigmoid', name='attention_vec')(lstm)#求解Attention权重\n",
    "attention=Activation('softmax',name='attention_weight')(attention)\n",
    "model=Multiply()([lstm, attention])#attention与LSTM对应数值相乘\n",
    "outputs = Dense(1, activation='relu')(model)\n",
    "model = Model(inputs=inputs, outputs=outputs)\n",
    "model.compile(loss='mse',optimizer='adam',metrics=['accuracy'])\n",
    "model.summary()#展示模型结构"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "52/52 [==============================] - 0s 6ms/step - loss: 18865.0918 - accuracy: 0.0000e+00 - val_loss: 9189.5625 - val_accuracy: 0.0000e+00\n",
      "Epoch 2/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 18585.4805 - accuracy: 0.0000e+00 - val_loss: 8918.0283 - val_accuracy: 0.0000e+00\n",
      "Epoch 3/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 17584.8906 - accuracy: 0.0000e+00 - val_loss: 7949.5532 - val_accuracy: 0.0000e+00\n",
      "Epoch 4/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 6730.7002 - accuracy: 0.0000e+00 - val_loss: 2582.2476 - val_accuracy: 0.0000e+00\n",
      "Epoch 5/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 2893.6785 - accuracy: 0.0000e+00 - val_loss: 1587.0033 - val_accuracy: 0.0000e+00\n",
      "Epoch 6/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 2220.4319 - accuracy: 0.0000e+00 - val_loss: 1767.8275 - val_accuracy: 0.0000e+00\n",
      "Epoch 7/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 1819.6636 - accuracy: 0.0000e+00 - val_loss: 2156.7981 - val_accuracy: 0.0000e+00\n",
      "Epoch 8/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 1613.7834 - accuracy: 0.0000e+00 - val_loss: 2004.3506 - val_accuracy: 0.0000e+00\n",
      "Epoch 9/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 1291.7950 - accuracy: 0.0000e+00 - val_loss: 1743.4725 - val_accuracy: 0.0000e+00\n",
      "Epoch 10/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 1154.0035 - accuracy: 0.0000e+00 - val_loss: 1535.7778 - val_accuracy: 0.0000e+00\n",
      "Epoch 11/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 882.7735 - accuracy: 0.0000e+00 - val_loss: 1890.7881 - val_accuracy: 0.0000e+00\n",
      "Epoch 12/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 759.8930 - accuracy: 0.0000e+00 - val_loss: 1261.8778 - val_accuracy: 0.0000e+00\n",
      "Epoch 13/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 629.3291 - accuracy: 0.0000e+00 - val_loss: 1341.7013 - val_accuracy: 0.0000e+00\n",
      "Epoch 14/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 645.2457 - accuracy: 0.0000e+00 - val_loss: 1137.4673 - val_accuracy: 0.0000e+00\n",
      "Epoch 15/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 588.3474 - accuracy: 0.0000e+00 - val_loss: 1115.6135 - val_accuracy: 0.0000e+00\n",
      "Epoch 16/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 547.8922 - accuracy: 0.0000e+00 - val_loss: 1021.8643 - val_accuracy: 0.0000e+00\n",
      "Epoch 17/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 449.3040 - accuracy: 0.0000e+00 - val_loss: 1076.5472 - val_accuracy: 0.0000e+00\n",
      "Epoch 18/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 447.7881 - accuracy: 0.0000e+00 - val_loss: 992.8433 - val_accuracy: 0.0000e+00\n",
      "Epoch 19/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 459.7743 - accuracy: 0.0000e+00 - val_loss: 832.2431 - val_accuracy: 0.0000e+00\n",
      "Epoch 20/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 445.7362 - accuracy: 0.0000e+00 - val_loss: 1042.5281 - val_accuracy: 0.0000e+00\n",
      "Epoch 21/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 469.2104 - accuracy: 0.0000e+00 - val_loss: 802.7836 - val_accuracy: 0.0000e+00\n",
      "Epoch 22/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 467.6004 - accuracy: 0.0000e+00 - val_loss: 849.7312 - val_accuracy: 0.0000e+00\n",
      "Epoch 23/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 446.4549 - accuracy: 0.0000e+00 - val_loss: 875.8225 - val_accuracy: 0.0000e+00\n",
      "Epoch 24/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 419.9922 - accuracy: 0.0000e+00 - val_loss: 787.7057 - val_accuracy: 0.0000e+00\n",
      "Epoch 25/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 415.9788 - accuracy: 0.0000e+00 - val_loss: 733.2205 - val_accuracy: 0.0000e+00\n",
      "Epoch 26/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 365.4932 - accuracy: 0.0000e+00 - val_loss: 693.4637 - val_accuracy: 0.0000e+00\n",
      "Epoch 27/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 363.7121 - accuracy: 0.0000e+00 - val_loss: 712.4803 - val_accuracy: 0.0000e+00\n",
      "Epoch 28/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 331.6456 - accuracy: 0.0000e+00 - val_loss: 697.0228 - val_accuracy: 0.0000e+00\n",
      "Epoch 29/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 321.3102 - accuracy: 0.0000e+00 - val_loss: 668.8275 - val_accuracy: 0.0000e+00\n",
      "Epoch 30/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 294.2795 - accuracy: 0.0000e+00 - val_loss: 673.9464 - val_accuracy: 0.0000e+00\n",
      "Epoch 31/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 287.1324 - accuracy: 0.0000e+00 - val_loss: 670.3339 - val_accuracy: 0.0000e+00\n",
      "Epoch 32/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 281.0581 - accuracy: 0.0000e+00 - val_loss: 691.5383 - val_accuracy: 0.0000e+00\n",
      "Epoch 33/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 298.7249 - accuracy: 0.0000e+00 - val_loss: 705.2952 - val_accuracy: 0.0000e+00\n",
      "Epoch 34/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 305.3274 - accuracy: 0.0000e+00 - val_loss: 677.0185 - val_accuracy: 0.0000e+00\n",
      "Epoch 35/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 265.6088 - accuracy: 0.0000e+00 - val_loss: 668.2397 - val_accuracy: 0.0000e+00\n",
      "Epoch 36/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 295.2708 - accuracy: 0.0000e+00 - val_loss: 646.6524 - val_accuracy: 0.0000e+00\n",
      "Epoch 37/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 257.5055 - accuracy: 0.0000e+00 - val_loss: 591.5176 - val_accuracy: 0.0000e+00\n",
      "Epoch 38/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 269.3846 - accuracy: 0.0000e+00 - val_loss: 603.2484 - val_accuracy: 0.0000e+00\n",
      "Epoch 39/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 265.8794 - accuracy: 0.0000e+00 - val_loss: 652.7669 - val_accuracy: 0.0000e+00\n",
      "Epoch 40/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 270.7534 - accuracy: 0.0000e+00 - val_loss: 641.6552 - val_accuracy: 0.0000e+00\n",
      "Epoch 41/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 267.4128 - accuracy: 0.0000e+00 - val_loss: 608.4535 - val_accuracy: 0.0000e+00\n",
      "Epoch 42/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 258.2084 - accuracy: 0.0000e+00 - val_loss: 608.1451 - val_accuracy: 0.0000e+00\n",
      "Epoch 43/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 245.2621 - accuracy: 0.0000e+00 - val_loss: 547.3765 - val_accuracy: 0.0000e+00\n",
      "Epoch 44/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 224.6637 - accuracy: 0.0000e+00 - val_loss: 537.5975 - val_accuracy: 0.0000e+00\n",
      "Epoch 45/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 248.7733 - accuracy: 0.0000e+00 - val_loss: 495.7101 - val_accuracy: 0.0000e+00\n",
      "Epoch 46/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 223.2580 - accuracy: 0.0000e+00 - val_loss: 484.5293 - val_accuracy: 0.0000e+00\n",
      "Epoch 47/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 221.5629 - accuracy: 0.0000e+00 - val_loss: 499.7228 - val_accuracy: 0.0000e+00\n",
      "Epoch 48/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 213.4862 - accuracy: 0.0000e+00 - val_loss: 510.2286 - val_accuracy: 0.0000e+00\n",
      "Epoch 49/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 222.3028 - accuracy: 0.0000e+00 - val_loss: 481.2324 - val_accuracy: 0.0000e+00\n",
      "Epoch 50/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 211.0124 - accuracy: 0.0000e+00 - val_loss: 507.8706 - val_accuracy: 0.0000e+00\n",
      "Epoch 51/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 233.7793 - accuracy: 0.0000e+00 - val_loss: 503.4849 - val_accuracy: 0.0000e+00\n",
      "Epoch 52/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 222.3664 - accuracy: 0.0000e+00 - val_loss: 570.3406 - val_accuracy: 0.0000e+00\n",
      "Epoch 53/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 225.9970 - accuracy: 0.0000e+00 - val_loss: 517.7255 - val_accuracy: 0.0000e+00\n",
      "Epoch 54/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "52/52 [==============================] - 0s 2ms/step - loss: 213.7133 - accuracy: 0.0000e+00 - val_loss: 513.2588 - val_accuracy: 0.0000e+00\n",
      "Epoch 55/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 194.9582 - accuracy: 0.0000e+00 - val_loss: 514.1225 - val_accuracy: 0.0000e+00\n",
      "Epoch 56/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 204.3272 - accuracy: 0.0000e+00 - val_loss: 486.3162 - val_accuracy: 0.0000e+00\n",
      "Epoch 57/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 193.4297 - accuracy: 0.0000e+00 - val_loss: 457.2967 - val_accuracy: 0.0000e+00\n",
      "Epoch 58/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 181.3664 - accuracy: 0.0000e+00 - val_loss: 453.2860 - val_accuracy: 0.0000e+00\n",
      "Epoch 59/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 217.1401 - accuracy: 0.0000e+00 - val_loss: 558.2360 - val_accuracy: 0.0000e+00\n",
      "Epoch 60/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 233.7190 - accuracy: 0.0000e+00 - val_loss: 571.1138 - val_accuracy: 0.0000e+00\n",
      "Epoch 61/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 215.1942 - accuracy: 0.0000e+00 - val_loss: 553.8860 - val_accuracy: 0.0000e+00\n",
      "Epoch 62/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 252.4379 - accuracy: 0.0000e+00 - val_loss: 473.5533 - val_accuracy: 0.0000e+00\n",
      "Epoch 63/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 201.2331 - accuracy: 0.0000e+00 - val_loss: 494.4694 - val_accuracy: 0.0000e+00\n",
      "Epoch 64/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 184.4893 - accuracy: 0.0000e+00 - val_loss: 485.0182 - val_accuracy: 0.0000e+00\n",
      "Epoch 65/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 227.2511 - accuracy: 0.0000e+00 - val_loss: 537.7515 - val_accuracy: 0.0000e+00\n",
      "Epoch 66/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 262.8288 - accuracy: 0.0000e+00 - val_loss: 511.9914 - val_accuracy: 0.0000e+00\n",
      "Epoch 67/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 217.9164 - accuracy: 0.0000e+00 - val_loss: 496.9816 - val_accuracy: 0.0000e+00\n",
      "Epoch 68/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 193.9487 - accuracy: 0.0000e+00 - val_loss: 514.4773 - val_accuracy: 0.0000e+00\n",
      "Epoch 69/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 208.1967 - accuracy: 0.0000e+00 - val_loss: 476.0217 - val_accuracy: 0.0000e+00\n",
      "Epoch 70/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 187.0381 - accuracy: 0.0000e+00 - val_loss: 451.6778 - val_accuracy: 0.0000e+00\n",
      "Epoch 71/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 179.6617 - accuracy: 0.0000e+00 - val_loss: 408.5055 - val_accuracy: 0.0000e+00\n",
      "Epoch 72/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 165.5068 - accuracy: 0.0000e+00 - val_loss: 444.5580 - val_accuracy: 0.0000e+00\n",
      "Epoch 73/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 172.8876 - accuracy: 0.0000e+00 - val_loss: 410.4533 - val_accuracy: 0.0000e+00\n",
      "Epoch 74/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 162.2940 - accuracy: 0.0000e+00 - val_loss: 376.4404 - val_accuracy: 0.0000e+00\n",
      "Epoch 75/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 153.5679 - accuracy: 0.0000e+00 - val_loss: 415.0610 - val_accuracy: 0.0000e+00\n",
      "Epoch 76/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 157.8506 - accuracy: 0.0000e+00 - val_loss: 402.0456 - val_accuracy: 0.0000e+00\n",
      "Epoch 77/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 149.5378 - accuracy: 0.0000e+00 - val_loss: 404.5312 - val_accuracy: 0.0000e+00\n",
      "Epoch 78/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 148.6674 - accuracy: 0.0000e+00 - val_loss: 391.0363 - val_accuracy: 0.0000e+00\n",
      "Epoch 79/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 167.8937 - accuracy: 0.0000e+00 - val_loss: 412.0544 - val_accuracy: 0.0000e+00\n",
      "Epoch 80/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 164.7576 - accuracy: 0.0000e+00 - val_loss: 403.4123 - val_accuracy: 0.0000e+00\n",
      "Epoch 81/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 165.6757 - accuracy: 0.0000e+00 - val_loss: 416.1315 - val_accuracy: 0.0000e+00\n",
      "Epoch 82/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 149.2726 - accuracy: 0.0000e+00 - val_loss: 358.0013 - val_accuracy: 0.0000e+00\n",
      "Epoch 83/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 141.7615 - accuracy: 0.0000e+00 - val_loss: 391.7559 - val_accuracy: 0.0000e+00\n",
      "Epoch 84/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 145.0753 - accuracy: 0.0000e+00 - val_loss: 385.5498 - val_accuracy: 0.0000e+00\n",
      "Epoch 85/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 148.4872 - accuracy: 0.0000e+00 - val_loss: 381.2675 - val_accuracy: 0.0000e+00\n",
      "Epoch 86/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 139.7431 - accuracy: 0.0000e+00 - val_loss: 372.5687 - val_accuracy: 0.0000e+00\n",
      "Epoch 87/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 134.4298 - accuracy: 0.0000e+00 - val_loss: 382.2878 - val_accuracy: 0.0000e+00\n",
      "Epoch 88/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 135.5885 - accuracy: 0.0000e+00 - val_loss: 366.1591 - val_accuracy: 0.0000e+00\n",
      "Epoch 89/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 131.2897 - accuracy: 0.0000e+00 - val_loss: 414.3078 - val_accuracy: 0.0000e+00\n",
      "Epoch 90/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 133.9782 - accuracy: 0.0000e+00 - val_loss: 394.3565 - val_accuracy: 0.0000e+00\n",
      "Epoch 91/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 141.5120 - accuracy: 0.0000e+00 - val_loss: 407.9890 - val_accuracy: 0.0000e+00\n",
      "Epoch 92/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 137.7868 - accuracy: 0.0000e+00 - val_loss: 339.7355 - val_accuracy: 0.0000e+00\n",
      "Epoch 93/100\n",
      "52/52 [==============================] - 0s 3ms/step - loss: 141.6196 - accuracy: 0.0000e+00 - val_loss: 331.5241 - val_accuracy: 0.0000e+00\n",
      "Epoch 94/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 135.3543 - accuracy: 0.0000e+00 - val_loss: 445.2190 - val_accuracy: 0.0000e+00\n",
      "Epoch 95/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 149.9743 - accuracy: 0.0000e+00 - val_loss: 351.2598 - val_accuracy: 0.0000e+00\n",
      "Epoch 96/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 133.8950 - accuracy: 0.0000e+00 - val_loss: 344.1158 - val_accuracy: 0.0000e+00\n",
      "Epoch 97/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 140.2363 - accuracy: 0.0000e+00 - val_loss: 388.9391 - val_accuracy: 0.0000e+00\n",
      "Epoch 98/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 135.8693 - accuracy: 0.0000e+00 - val_loss: 307.6921 - val_accuracy: 0.0000e+00\n",
      "Epoch 99/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 124.4345 - accuracy: 0.0000e+00 - val_loss: 316.3007 - val_accuracy: 0.0000e+00\n",
      "Epoch 100/100\n",
      "52/52 [==============================] - 0s 2ms/step - loss: 133.7793 - accuracy: 0.0000e+00 - val_loss: 325.1414 - val_accuracy: 0.0000e+00\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "如果直接加载模型，则不需要运行这一段训练\n",
    "'''\n",
    "#训练模型\n",
    "history=model.fit(x_train, y_train, epochs = epoch, batch_size = batch_size,\n",
    "                  shuffle=False,validation_split=.1) #训练模型epoch次\n",
    "model.save(save_model)#保存模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#加载模型\n",
    "from tensorflow.keras.models import load_model\n",
    "model=load_model(save_model)\n",
    "#获得网络权重\n",
    "#weights = np.array(model.get_weights())\n",
    "#print(weights)\n",
    "#输出attention层权重\n",
    "attention_layer_model = Model(inputs=model.input,outputs=model.get_layer('attention_weight').output)\n",
    "attention_weight = attention_layer_model.predict(x_train)\n",
    "attention_weight_final=np.mean(np.array(attention_weight), axis=0)\n",
    "pd.DataFrame(attention_weight_final, columns=['attention (%)']).plot(kind='bar',\n",
    "                                                                         title='Attention Mechanism as '\n",
    "                                                                               'a function of input'\n",
    "                                                                               ' dimensions.')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#迭代图像\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "epochs_range = range(epoch)\n",
    "plt.plot(epochs_range, loss, label='Train Loss')\n",
    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title('Train and Val Loss')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "训练集上的MAE/RMSE/MAPE/涨跌准确率\n",
      "7.565399707244535\n",
      "11.795112684116665\n",
      "14.052647557953474\n",
      "0.7812756638379414\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#模型进行检验\n",
    "def mape(y_true, y_pred):\n",
    "    return np.mean(np.abs((y_pred - y_true) / y_true)) * 100\n",
    "def up_down_accuracy(y_true, y_pred):\n",
    "    y_var_test=y_true[1:]-y_true[:len(y_true)-1]#实际涨跌\n",
    "    y_var_predict=y_pred[1:]-y_pred[:len(y_pred)-1]#原始涨跌\n",
    "    txt=np.zeros(len(y_var_test))\n",
    "    for i in range(len(y_var_test-1)):#计算数量\n",
    "        txt[i]=np.sign(y_var_test[i])==np.sign(y_var_predict[i])\n",
    "    result=sum(txt)/len(txt)\n",
    "    return result\n",
    "#在训练集上的拟合结果\n",
    "y_train_predict=model.predict(x_train)*scale\n",
    "y_train_predict=y_train_predict[:,0]\n",
    "draw=pd.concat([pd.DataFrame(y_train),pd.DataFrame(y_train_predict)],axis=1)\n",
    "draw.iloc[200:500,0].plot(figsize=(12,6))\n",
    "draw.iloc[200:500,1].plot(figsize=(12,6))\n",
    "plt.legend(('real', 'predict'),fontsize='15')\n",
    "plt.title(\"Train Data\",fontsize='30') #添加标题\n",
    "#输出结果\n",
    "print('训练集上的MAE/RMSE/MAPE/涨跌准确率')\n",
    "print(mean_absolute_error(y_train_predict, y_train))\n",
    "print(np.sqrt(mean_squared_error(y_train_predict, y_train)))\n",
    "print(mape(y_train_predict, y_train) )\n",
    "print(up_down_accuracy(y_train_predict,y_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "测试集上的MAE/RMSE/MAPE/涨跌准确率\n",
      "8.26394595092286\n",
      "12.78607221554969\n",
      "11.36395165555698\n",
      "0.7413793103448276\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#在测试集上的预测\n",
    "y_test_predict=model.predict(x_test)*scale\n",
    "y_test_predict=y_test_predict[:,0]\n",
    "draw=pd.concat([pd.DataFrame(y_test),pd.DataFrame(y_test_predict)],axis=1);\n",
    "draw.iloc[:,0].plot(figsize=(12,6))\n",
    "draw.iloc[:,1].plot(figsize=(12,6))\n",
    "plt.legend(('real', 'predict'),loc='upper right',fontsize='15')\n",
    "plt.title(\"Test Data\",fontsize='30') #添加标题\n",
    "#输出结果\n",
    "print('测试集上的MAE/RMSE/MAPE/涨跌准确率')\n",
    "print(mean_absolute_error(y_test_predict, y_test))\n",
    "print(np.sqrt(mean_squared_error(y_test_predict, y_test)))\n",
    "print(mape(y_test_predict,  y_test) )\n",
    "print(up_down_accuracy(y_test_predict,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}