时序预测pytorch

时序预测是指基于历史数据来预测未来时间点的数值或趋势。在PyTorch中，可以使用神经网络模型进行时序预测。以下是一个简单的PyTorch时序预测的示例代码：

import torch
import torch.nn as nn

# 准备数据
data = torch.tensor([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0])
lookback = 3
input_size = 1
output_size = 1
x = torch.zeros(len(data) - lookback, lookback, input_size)
y = torch.zeros(len(data) - lookback, output_size)
for i in range(len(data) - lookback):
    x[i] = data[i:i+lookback].view(-1, 1)
    y[i] = data[i+lookback:i+lookback+output_size].view(-1, 1)

# 定义模型
class LSTM(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(LSTM, self).__init__()
        self.lstm = nn.LSTM(input_size, hidden_size)
        self.linear = nn.Linear(hidden_size, output_size)

    def forward(self, x):
        output, _ = self.lstm(x)
        output = self.linear(output[:, -1])
        return output

model = LSTM(input_size, 16, output_size)

# 定义损失函数和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr=0.01)

# 训练模型
num_epochs = 100
for epoch in range(num_epochs):
    y_pred = model(x)
    loss = criterion(y_pred, y)
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()
    if (epoch+1) % 10 == 0:
        print('Epoch [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))

# 预测未来值
with torch.no_grad():
    future = 5
    pred = data[-lookback:].view(-1, 1)
    for i in range(future):
        input = pred[-lookback:].view(1, lookback, input_size)
        output = model(input)
        pred = torch.cat((pred, output.view(-1, 1)), dim=0)
    print('Predicted future values:', pred[lookback:])

在这个示例中，我们使用了一个LSTM模型进行时序预测。首先，我们将历史数据拆分成输入和输出，然后定义LSTM模型，损失函数和优化器。接下来，我们训练模型并预测未来值。