特征縮放：

1歸一化
X-Xmin/(Xmax-Xmin) (0,1)
2標(biāo)準(zhǔn)化
X-均值/期望

7個時間步 最后預(yù)測第八天的收盤價格 即圖中的y

def forward(self, x):# Initialize hidden and cell statesh_0 = Variable(torch.zeros(self.num_layers, x.size(0), self.hidden_size))# 上一個時間步的預(yù)測值c_0 = Variable(torch.zeros(self.num_layers, x.size(0), self.hidden_size))# 記憶單元 h0 = sigmod(c0)'''送進來的每條樣本都生成記憶單元向 量(層數(shù)，樣本數(shù)量，記憶單元長度)（1,724*0.7,5）'''# Propagate input through LSTM_, (h_out, _) = self.lstm(x, (h_0, c_0)) # 取得最后的輸出結(jié)果 # 此處分別是 y1-y7,中間結(jié)果y7,要送入下一步的記憶單元ch_out = h_out.view(-1, self.hidden_size)# 行列變換 （n條樣本，5） 每個樣本的特征數(shù)是5 out = self.fc(h_out)# 連接一個全連接層 輸出最后的結(jié)果h7return out'''# Propagate input through LSTM_, (h_out, _) = self.lstm(x, (h_0, c_0)) # 取得最后的輸出結(jié)果h_out = h_out.view(-1, self.hidden_size)out = self.fc(h_out)return out

''' This script shows how to predict stock prices using a basic RNN ''' import torch import torch.nn as nn from torch.autograd import Variable import numpy as nptorch.manual_seed(777) # reproducibilityimport matplotlib.pyplot as plt# 特征縮放 歸一化 def MinMaxScaler(data):numerator = data - np.min(data, 0)denominator = np.max(data, 0) - np.min(data, 0)# noise term prevents the zero divisionreturn numerator / (denominator + 1e-7)# train Parameters learning_rate = 0.01 num_epochs = 500 input_size = 5 hidden_size = 5 num_classes = 1 timesteps = seq_length = 7 num_layers = 1 # number of layers in RNN# Open, High, Low, Volume, Close xy = np.loadtxt('data-02-stock_daily.csv', delimiter=',') xy = xy[::-1] # reverse order (chronically ordered) 拿出所有行除了最后一列 特征部分 xy = MinMaxScaler(xy) x = xy y = xy[:, [-1]] # Close as label 拿到最后一列# build a dataset dataX = [] dataY = [] for i in range(0, len(y) - seq_length):_x = x[i:i + seq_length] # [0,7） x0-x6 _x是一條樣本的特征_y = y[i + seq_length] # Next close price 最后一列收盤價格 即y[7] _y是一條樣本的標(biāo)簽 print(_x, "->", _y)dataX.append(_x)dataY.append(_y)# train/test split train_size = int(len(dataY) * 0.7) #724條數(shù)據(jù) *0.7 test_size = len(dataY) - train_size #724*0.3 trainX = torch.Tensor(np.array(dataX[0:train_size])) trainX = Variable(trainX)# 張量變自動求導(dǎo)的過程 testX = torch.Tensor(np.array(dataX[train_size:len(dataX)])) testX = Variable(testX) trainY = torch.Tensor(np.array(dataY[0:train_size])) trainY = Variable(trainY) testY = torch.Tensor(np.array(dataY[train_size:len(dataY)])) testY = Variable(testY)class LSTM(nn.Module):# 分類問題只有一個數(shù)，一次性送入5條樣本數(shù)據(jù)，記憶單元向量的長度為5,一個rnn單元def __init__(self, num_classes, input_size, hidden_size, num_layers):super(LSTM, self).__init__()self.num_classes = num_classesself.num_layers = num_layersself.input_size = input_sizeself.hidden_size = hidden_sizeself.seq_length = seq_length# Set parameters for RNN block# Note: batch_first=False by default.# When true, inputs are (batch_size, sequence_length, input_dimension)# instead of (sequence_length, batch_size, input_dimension)self.lstm = nn.LSTM(input_size=input_size, hidden_size=hidden_size,num_layers=num_layers, batch_first=True)# Fully connected layerself.fc = nn.Linear(hidden_size, num_classes)def forward(self, x):# Initialize hidden and cell statesh_0 = Variable(torch.zeros(self.num_layers, x.size(0), self.hidden_size))# 上一個時間步的預(yù)測值c_0 = Variable(torch.zeros(self.num_layers, x.size(0), self.hidden_size))# 記憶單元 h0 = sigmod(c0)'''送進來的每條樣本都生成記憶單元向 量(層數(shù)，樣本數(shù)量，記憶單元長度)（1,724*0.7,5）'''# Propagate input through LSTM_, (h_out, _) = self.lstm(x, (h_0, c_0)) # 取得最后的輸出結(jié)果 # 此處分別是 y1-y7,中間結(jié)果y7,要送入下一步的記憶單元ch_out = h_out.view(-1, self.hidden_size)# 行列變換 （n條樣本，5） 每個樣本的特征數(shù)是5 out = self.fc(h_out)# 連接一個全連接層 輸出最后的結(jié)果h7return out# Instantiate RNN model lstm = LSTM(num_classes, input_size, hidden_size, num_layers)# Set loss and optimizer function criterion = torch.nn.MSELoss() # mean-squared error for regression optimizer = torch.optim.Adam(lstm.parameters(), lr=learning_rate)# Train the model for epoch in range(num_epochs):outputs = lstm(trainX)# 一次性把所有訓(xùn)練樣本都輸入進來 沒有一批批用batchsize送數(shù)據(jù)optimizer.zero_grad()# 對每個epcho 偏導(dǎo)數(shù)置0 防止持續(xù)累積 # obtain the loss functionloss = criterion(outputs, trainY)loss.backward()# 計算偏導(dǎo)數(shù)optimizer.step()# 更新參數(shù)print("Epoch: %d, loss: %1.5f" % (epoch, loss.item()))print("Learning finished!")# Test the model lstm.eval()# 關(guān)閉訓(xùn)練技巧 t est_predict = lstm(testX)# Plot predictions test_predict = test_predict.data.numpy() testY = testY.data.numpy() plt.plot(testY,c='y') plt.plot(test_predict,c='b') print(testY.shape) print(test_predict.shape) plt.xlabel("Time Period") plt.ylabel("Stock Price") plt.show()

總結(jié)

以上是生活随笔為你收集整理的LSTM股票价格预测的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。