Add hf_hub_download #2
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
facebook-github-bot
added
the
CLA Signed
Member
|
Thank you Niels!! This is awesome. I will need to look into whether we are allowed to upload the native PyTorch weights to our HuggingFace repo or if we have to create a new repo |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hi,
Niels here from the open-source team at HF. It's great to see native HF Transformers support for the models!
This PR showcases that you could easily also host the original PyTorch checkpoints on 🤗.
This has the advantage that people don't need to download a checkpoint themselves anymore, it just downloads automatically from the hub.
This is also done for other works, like SigLIP-2 for example, where both the Transformers-compatible checkpoints as well as the original Jax checkpoints are on the hub.