Szymon Maszke

Will check out `HDF5` and other formats and update for upcoming `0.3.0` release. If you have any suggestions feel free to post, thanks.

Thanks for the suggestion, if you wish to implement this feature feel free to open the PR!

See [PyTorch's Feature Request](https://github.com/pytorch/pytorch/issues/3867), in case of `torchlayers` there are also some quirks one would have to solve as well. `same` padding for even values would have to be done...

Yes, S3 is possible but you would have to build your own `torchlambda` Docker image (see [`torchlambda build` command documentation](https://github.com/szymonmaszke/torchlambda/wiki/Commands#torchlambda-build), specifically `--aws-components "s3"` should be passed during building). Also you...

Yes, you could do that, as `torchlambda` consumes PyTorch's C++ frontend so you can do most of the things it supports, but might have to get your hands dirty with...

Thanks for catching, will improve upon this functionality and close as soon as I'm done with it.

I think all the cores should be used out of the box even with static build. You may also try to change some PyTorch flags as described in [documentation](https://github.com/szymonmaszke/torchlambda/wiki/Commands#optional-arguments-2) and...

Yes it should be fairly easy with `.yaml` settings. Will add a comment with it tomorrow. BTW. Usually you don't need to delve into C++ with this tool.

Update: `base64` return type is not currently supported, you would have to create C++ on your own based on `torchlambda template` generated setup.

For now, you can see how to modify C++ source code [in this issue](https://github.com/szymonmaszke/torchlambda/issues/5), transforming to `base64` needs some knowledge of C++ and AWS API & PyTorch C++ API though.