8bitmp3

@mikowals I noticed that examples in the `swift-models` repo, such as the `BERT-Cola` and others, set the calculations to be run on an accelerator via XLA on the X10 backend...

@BenLucas7 @dan-zheng Take a look at this https://stackoverflow.com/questions/11530090/adding-a-new-entry-to-the-path-variable-in-zsh Having migrated to ZSH with Catalina macOS, you may stumble upon a few issues when setting variables the old way. Can you...

@BenLucas7 @dan-zheng Wonder if it has to do with macOS accounts created pre-Catalina and therefore still relying on bash 3.2 even after enabling ZSH (`%`). There's a good write-up on...

@BenLucas7 @dan-zheng judging by your output ``` $ ~ % swiftenv versions * tensorflow-RELEASE-0.9 5.2.4 $ ~ % swiftenv global tensorflow-RELEASE-0.9 ``` You set v0.9 as `global` and the ran...

@BenLucas7 @dan-zheng Ok, I got a similar issue, maybe it's not ZSH. - `swiftenv` has been installed. - Ran `xcode-select --install` in the terminal to reregister the toolkit. - Deleted...

Thanks @ematejska. Without the read eval print loop support, what do you and the team suggest we do to have CLI support here when using `swiftenv`?

Since the site is built with MkDocs `mkdocs-material`, do you accept improvement PRs? I'm familiar with it. I also have some Flax and TensorFlow docs experience. Awesome work on `elegy`

Can't speak for the team but have you seen https://github.com/deepmind/jraph? (copying @jg8610).

@sudo-michael @wookayin Check out the new Checkpointing guide by @IvyZX https://github.com/google/flax/blob/main/docs/guides/use_checkpointing.ipynb

PTAL @MarkDaoust @markmcd