Change DataLoader Collate Function

[ehofesmann]

Currently for datasets with bounding boxes, we need to specify the max bounding boxes possible so all output batches are of the same size:

ViP/datasets/ImageNetVID.py

Line 27 in 74776f2

self.max_objects = 38

What we should do is use a custom collate function in the DataLoader like used in the Pytorch detection tutorial:

https://github.com/pytorch/vision/blob/6c2cda6a0eda4c835f96f18bb2b3be5043d96ad2/references/detection/utils.py#L237

https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html

[natlouis]

Yeah collate is very useful for zero-padding your samples when there's a different number of objects (or lengths in other cases).
But in my experience, it's specific to each problem. I'm not sure how it could generalize across all datasets if we add that to the DataLoader.