Padded sequences?

[drozzy]

Just curious if this works with padded sequences?
https://pytorch.org/docs/stable/generated/torch.nn.utils.rnn.pad_sequence.html

I have sequences of variable length and I would like to use torchdyn for classification.
Currently I am doing windowing, but I would rather use padding.

[Zymrael]

Hi @drozzy, could you provide a more specific example of what you're trying to do? There are several models in torchdyn, developed for different tasks. Although we have not designed any component with padded seqs in mind, it might be easy to use them with some of these architectures, depending on your objective.

Which torchdyn classes does your classifier utilize?

[drozzy]

Maybe my question is wrong. I'm basically working with variable length sequences and I want to do classification using torchdyn.

From torchdyn I'm just using something like:
model = NeuralDE(f, sensitivity='adjoint', solver='dopri5').to(device)

[drozzy]

I am not sure what I was asking. I'll close it and re-open it if I figure it out!