SparseStorage tensor.Long assert

[Pendulun]

Hi, thank you for your time on developing this project. I have a question that I think is rather simple: Why do you have to enforce that the SparseStorage's col and other params to be tensor.Long? Is it really necessary? Thanks

[rusty1s]

Currently, that is necessary. The reason for that is that a lot of functionality in PyTorch assumes indices to be of type torch.long, e.g., torch.index_select() only works with torch.long indices. I heard that PyTorch team is actively working on more diverse types for these operators in recent versions, so one day this shouldn't be necessary anymore.

[Pendulun]

Thanks, I'm in a situation where forcing me to use LongTensors is really a waste of memory (IntTensor is enough) that's why the question. Looking forward to the mentioned work of the PyTorch team.

[github-actions]

This issue had no activity for 6 months. It will be closed in 2 weeks unless there is some new activity. Is this issue already resolved?