rygrrg.txt

import pandas as pd
import numpy as np
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense, LSTM
import matplotlib.pyplot as plt
from tensorflow.keras.optimizers import Adam
from sklearn.metrics import mean_squared_error



data = pd.read_csv('./data.tsv',sep='\t',skipinitialspace=True)
data.head()

features = data.iloc[:, :18].values
labels = data.iloc[:, 18].values

scaler = MinMaxScaler(feature_range=(0, 1))
features_scaled = scaler.fit_transform(features)
labels_scaled = labels.reshape(-1,1)



train_size = int(len(features_scaled) * 0.8)#总行数的70%为训练集
val_size = int(len(features_scaled) * 0.15)#10%为验证集
test_size = len(features_scaled) - train_size - val_size#剩下20%为测试集
train_features = features_scaled[:train_size, :]
train_labels = labels_scaled[:train_size]
val_features = features_scaled[train_size:train_size+val_size, :]
val_labels = labels_scaled[train_size:train_size+val_size]
test_features = features_scaled[train_size+val_size:, :]
test_labels = labels_scaled[train_size+val_size:]



def create_lstm_dataset(features, labels, time_steps=1):
    dataX, dataY = [], []
    for i in range(len(features) - time_steps):
        a = features[i:(i + time_steps), :]
        dataX.append(a)
        dataY.append(labels[i + time_steps])
    return np.array(dataX), np.array(dataY)

time_steps = 2#可调
trainX, trainY = create_lstm_dataset(train_features, train_labels, time_steps)
valX, valY = create_lstm_dataset(val_features, val_labels, time_steps)
testX, testY = create_lstm_dataset(test_features, test_labels, time_steps)



model = Sequential()
model.add(LSTM(50,activation='sigmoid', input_shape=(time_steps, 18)))
model.add(Dense(1))
learning_rate = 0.001
optimizer = Adam(lr=learning_rate)
model.compile(loss='mean_squared_error', optimizer=optimizer)

history = model.fit(trainX, trainY, validation_data=(valX, valY), epochs=100, batch_size=64, verbose=1)
train_loss = history.history['loss']
val_loss = history.history['val_loss']



plt.plot(train_loss, label='Train Loss')
plt.plot(val_loss, label='Validation Loss')
plt.legend()
plt.show()


test_predict = model.predict(testX)

plt.plot(test_predict, label='Predicted')
plt.plot(testY, label='True')
plt.legend()
plt.show()

MSE = mean_squared_error(test_predict,testY)
print ("均方误差为：", MSE)