facebookresearch / facebookresearch/jepa
Share HF Model checkpoints to the Hugging Face Hub 🤗
- Dominant language
- Python
- Stars
- 4.1k
- Forks
- 422
- PR merge metrics
- No merged PRs in 30d
Description
Hi there,
Congratulations on such a brilliant release. I'm VB, leading the advocacy efforts for open source at Hugging Face. I saw that the model checkpoints are released as CDN URLs. It'd be great if you could upload the model checkpoints over the Hugging Face Hub.
These ckpts can be part of the Facebook org on the Hugging Face Hub: https://huggingface.co/facebook :)
Uploading model checkpoints over on Hugging Face Hub comes with a couple of advantages:
1. It increases the visibility of the model checkpoints on the Hub.
2. It makes it easier for people to download the different weights and track the number of downloads too.
3. It makes it surprisingly easy to version control weights as well.
4. Here's a quick guide explaining how to upload models over on the Hub: https://huggingface.co/docs/hub/en/models-uploading
In addition to this, you can add support for directly loading the model checkpoints via the huggingface_hub library as well: https://huggingface.co/docs/huggingface_hub/v0.16.3/en/guides/download (it's just 2 lines of code).
Do let me know if you need any assistance with this.
Cheers,
VB
Contributor guide
Research direction
Locate the existing CDN checkpoint URLs and the code path that loads them, then review the linked Hugging Face Hub upload and huggingface_hub download guides. Done means the released checkpoints are available in the Facebook Hugging Face organization and the project can load them through huggingface_hub.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- huggingface, python, pytorch
- Domain
- machine-learning, release
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 42/100