Vadim Kantorov

@justinchuby `dynamo_export` is being improved, but just some time ago it did not support export of GRU: https://github.com/pytorch/pytorch/issues/91439 What do you think on possibility of backporting of export complex-dtype ops...

Regarding TorchScript -> FX graph conversion, there seems a newly merged: - https://github.com/pytorch/pytorch/pull/126920

I also think it's preferable to have such option (and also env. var controlling this) as a workaround for: - https://github.com/tqdm/tqdm/issues/760 And sometimes in the file logs, the progress bar...

Trying another library https://clalancette.github.io/pycdlib/, I am getting the following (`README` gets printed correctly)!: ``` For the introductory information to TeX Live, see the directories readme-txt.dir (plain text files) or readme-html.dir/...

Well, for filenames it's the story of Joliet/RockRidge extensions and so forth. `L9660_DEBUG=1 ./tb9660 ../texlive2024-20240312.iso cat 'README.TXT'` does not print anything useful. `L9660_DEBUG=1 ./tb9660 ../texlive2024-20240312.iso cat 'README.;1'` does something non-trivial...

Haha :) maybe you could add in README a warning/notice on long file names / name casing / Joliet not supported? I also wonder if explicit int64_t type could help....

Related nn.LayerNorm docs issue: - https://github.com/pytorch/pytorch/issues/51455 (it would be nice to have a super-compact reference pytorch impl of layernorm in docs for it's much clearer on what's aggregated and even...

> To easiest way to get what we want in PyTorch right now is `F.layer_norm(x.permute(0, 2, 3, 1), self.normalized_shape, self.weight, self.bias, self.eps).permute(0, 3, 1, 2)` ... however those permutes aren't...

This could also simplify `minima`, so that it can be also used in easy way for nextjs or 11ty

I have the same error - but without any containers. What to do? The job won't start, and cannot even be stopped - always `pending`. I cannot even understand what's...