argument 'second' in '_FunctionCorrelation' is not contiguous if batch size bigger than 1

[pablopalafox]

Hi @sniklaus,

thanks a lot for your work! I managed to get the model overfit to one sample, using a batch size of 1. But when moving on to fine tuning the pretrained models and using a batch size different than 1, argument second in _FunctionCorrelation is not contiguous. I guess one option is to do second.contiguous(). But I was curious as to why this is happening. I've roughly spotted that function Backward returns a non-contiguous tensor (when batch size != 1). Let me know if this had happened to you before, or if maybe it's some issue from my side. Cheers!

EDIT: in particular, it's this line:

tensorMask = tensorOutput[:, -1:, :, :]

that is causing the output of Backward to not be contiguous when the batch size is not 1, which makes sense, since slicing can cause this (see this thread)

[sniklaus]

Are you triggering the following assert?

https://github.com/sniklaus/pytorch-pwc/blob/90c1ac66d5bc6ba1ce7f60064f128d519e946bea/correlation/correlation.py#L341

[sniklaus]

Closing due to inactivity, feel free to reopen in case this issue persists.

[pablopalafox]

Hi @sniklaus, this assert is being triggered.

I'm making the output of Backward contiguous as follows:

return (tensorOutput[:, :-1, :, :] * tensorMask).contiguous()

Seems to have solved my problem.

[sniklaus]

Thank you for sharing your findings, I would have suggested the same fix. :+1:

[pablopalafox]

Awesome, thanks a lot!

[eric-yyjau]

That's really helpful. Thank!