How to handle time series data tasks in PyTorch?

In PyTorch, handling time series data tasks typically requires using modules such as torch.nn.RNN, torch.nn.LSTM, torch.nn.GRU, as well as data loading tools like and

Here is a simple example demonstrating how to use PyTorch to handle a time series data task.

  1. Collection of data
import torch
from import Dataset

class TimeSeriesDataset(Dataset):
    def __init__(self, data): = data
    def __len__(self):
        return len(
    def __getitem__(self, idx):
  1. Define a model with RNN included.
import torch.nn as nn

class RNNModel(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers, output_size):
        super(RNNModel, self).__init__()
        self.rnn = nn.RNN(input_size, hidden_size, num_layers, batch_first=True)
        self.fc = nn.Linear(hidden_size, output_size)
    def forward(self, x):
        out, _ = self.rnn(x)
        out = self.fc(out[:, -1, :])
        return out
  1. Prepare the data and train the model.
# 定义超参数
input_size = 1
hidden_size = 64
num_layers = 1
output_size = 1
num_epochs = 100
learning_rate = 0.001

# 准备数据
data = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
dataset = TimeSeriesDataset(data)
dataloader = DataLoader(dataset, batch_size=1, shuffle=True)

# 初始化模型
model = RNNModel(input_size, hidden_size, num_layers, output_size)

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

# 训练模型
for epoch in range(num_epochs):
    for i, batch in enumerate(dataloader):
        inputs = batch.float().unsqueeze(2)
        targets = inputs.clone()
        outputs = model(inputs)
        loss = criterion(outputs, targets)
        if (i+1) % 10 == 0:
            print('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, i+1, len(dataloader), loss.item()))

In the example above, we first created a custom Dataset class to load time series data, then defined a model RNNModel containing an RNN, and finally prepared the data and trained the model. During training, we utilized mean square error loss function and Adam optimizer to optimize the model.

Leave a Reply 0

Your email address will not be published. Required fields are marked *