As you can see, it’s a bit intricate format but it’s needed this way. I am using PytorchLightning and in the .fit function, I receive the following error: each element in list of batch should be of equal size. If I use colllate_fn=lambda x:x the problem doesn’t appear but the format of my batch changes and I receive a list with all elements in the batch (len(batch) = batch_size).
Nevertheless, what I want is that each element of the list I show above have a dimension batch_size x d.
How can I do it? Cannot this be done with PyTorch?
It doesn’t work when I used PytorchLightning, I receive the mentioned error. Sorry, I should have specified, I thought it was something that would happen no matter what.