Zach Mueller
Zach Mueller
Agree with Amy here!
@DavidAfonsoValente can you rebase from main and push with `-f` (to not have the git commit history/diff all crazy) so we can double check tests are green? 🤗 Thanks!
cc @amyeroberts
@amyeroberts requesting a rereview after having to gut much of it on further testing 😅
FYI this will need https://github.com/huggingface/accelerate/pull/2631 as we need to upstream accelerate's ability to call train/eval on a wrapped optimizer
We've merged the accelerate portion in, so if anyone is trying this out in distributed fashions, you can do `pip install git+https://github.com/huggingface/accelerate` :)
@faaany are we sure that `model.device` *is a thing* across all these frameworks? At most I see `ModuleUtilsMixin` has `device` which is PyTorch specific (it gets added to `AutoModel`, but...
I couldn't recreate this warning. Can you try either: 1. Using the dev version of transformers `pip install git+https://github.com/huggingface/transformers` 2. Upgrading your PyTorch version `pip install torch -U` And let...
Hi all, after talking with the torch-xla team tensorflow is the issue here. You can try doing the following: ```bash pip uninstall tensorflow pip install tensorflow-cpu ```