co-separation
co-separation copied to clipboard
Variable number of objects
Hi, First of all, great project! I would like to ask about your dataset. it seems to me that each sample returned by the AudioVisualMUSICDataset's getitem method can be with variable number of objects(the "visuals" variable here: https://github.com/rhgao/co-separation/blob/bd4f4fd51f2d6090e1566d20d4e0d0c8d83dd842/data/audioVisual_dataset.py#L189). A question that arises from this is how the dataloader will batch these samples together. I can see in the attached link that you address this problem by padding with -1 the lables(here: https://github.com/rhgao/co-separation/blob/bd4f4fd51f2d6090e1566d20d4e0d0c8d83dd842/data/audioVisual_dataset.py#L183) but I did'nt figure it out where exactly you did this padding. Moreover, what about the variable objects problem mentioned above?
Thank you in advance