1dconv.py

import torch 
import torch.nn as nn
import torch.utils.data
from torch.autograd import Variable

# Minimal dataset
class Dataset(torch.utils.data.Dataset):
  def __len__(self):
    return 12

  def __getitem__(self, idx):
    return [idx, idx+1]

train_dataset = Dataset()
loader = torch.utils.data.DataLoader(dataset=train_dataset)
class CNN(nn.Module):
    def __init__(self):
        super(CNN, self).__init__()
        # Arguments should (by docs) be nn.Conv1d(#input channels, #output channels, kernel size)
        self.layer = nn.Conv1d(1, 1, 12)
        
    def forward(self, x):
        out = self.layer(x)
        return out
        
cnn = CNN()

for idx, (inputs, labels) in enumerate(loader):
  inputs = Variable(inputs)
  labels = Variable(labels)
  outputs = cnn(inputs)