from  keras.models import Sequential
from  keras.layers import Dense, Dropout, Embedding
from  keras.layers import InputLayer
from  keras.layers import LSTM
from  keras import backend as K
from keras.losses import mean_squared_error
from keras.layers.recurrent import SimpleRNN

import pymysql
import pickle
import os
import numpy

from data import load_pay_data

def mean_squared_error(y_true, y_pred):
    print(dir(y_true), y_true.consumers)
    print(y_true, y_pred)
    return K.mean(K.square(y_pred - y_true), axis=-1)

if __name__ == "__main__":

    x_train, y_train, tx_train, ty_train, input_shape = load_pay_data(80)

    
    model = Sequential()
    units = 500
    model.add(LSTM(units, activation='relu', dropout=0.1, input_shape=input_shape))
    
    # model.add(SimpleRNN(units, activation='relu'))
    # model.add(Dropout(0.1))
    model.add(Dense(1))
    model.summary()
    model.compile(loss = "mse", optimizer = 'adam')

    model.fit(x_train, y_train, batch_size=128, epochs=500)
    model.save("./predict_pay")

    p_data = model.predict(tx_train)
    for i in range(len(p_data)):
        print((p_data[i][0] - ty_train[i]) / ty_train[i], p_data[i][0], ty_train[i])
    # print("测结果:", p_data[i][0], "测:", tx_train[i], "真实:", ty_train[i])