-
Notifications
You must be signed in to change notification settings - Fork 376
[Refactor] Model Checklist Document #2043
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
docs/hub/model-release-checklist.md
Outdated
|
|
||
| ### Uploading Weights | ||
| - **Use separate repositories for different model weights**: | ||
| Create individual repositories for each variant of the same architecture. This lets you group them into a [collection](https://huggingface.co/docs/hub/en/collections), which users prefer when browsing many options. It also improves visibility because each model has its own URL (`hf.co/org/model-name`) and is easier to search. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| Create individual repositories for each variant of the same architecture. This lets you group them into a [collection](https://huggingface.co/docs/hub/en/collections), which users prefer when browsing many options. It also improves visibility because each model has its own URL (`hf.co/org/model-name`) and is easier to search. | |
| Create individual repositories for each variant of the same architecture. This lets you group them into a [collection](https://huggingface.co/docs/hub/en/collections), which are easier to navigate than directory listings. It also improves visibility because each model has its own URL (`hf.co/org/model-name`), makes search easier, and provides download counts for each one of your models. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
not sure if needed but we could also mention that it makes it easier to integrate with downstream libraries too.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Vaibhavs10 how do you mean? I thought downstream libraries only benefit from transformers integration. If you could guide me into understanding this point better I will add it here.
Vaibhavs10
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
very cool! thanks for the update, agree with most of the comments Pedro made and left a few suggestions.
docs/hub/model-release-checklist.md
Outdated
|
|
||
| ### Uploading Weights | ||
| - **Use separate repositories for different model weights**: | ||
| Create individual repositories for each variant of the same architecture. This lets you group them into a [collection](https://huggingface.co/docs/hub/en/collections), which users prefer when browsing many options. It also improves visibility because each model has its own URL (`hf.co/org/model-name`) and is easier to search. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
not sure if needed but we could also mention that it makes it easier to integrate with downstream libraries too.
docs/hub/model-release-checklist.md
Outdated
|
|
||
| Bonus: a recognised library also allows you to track downloads of your model over time. | ||
| > [!NOTE] | ||
| > A recognised library also allows you to track downloads of your model over time. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
maybe we could link this in the registered library bit: https://huggingface.co/docs/hub/en/models-adding-libraries
| ### Write a Comprehensive Model Card | ||
|
|
||
| ### Writing a Comprehensive Model Card | ||
| A well-crafted model card (the `README.md` in your repository) is essential for discoverability, reproducibility, and effective sharing. Include: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| A well-crafted model card (the `README.md` in your repository) is essential for discoverability, reproducibility, and effective sharing. Include: | |
| A well-crafted model card (the `README.md` in your repository) is essential for discoverability, reproducibility, and effective sharing. Make sure to cover: |
(there's another include in the following section)
| To make the process more seamless, click **Import model card template** to pre-fill the `README.md`s with placeholders. | ||
|
|
||
| 1. **Library Integration**: If possible, add support for one of the many [libraries integrated with the Hugging Face Hub](https://huggingface.co/docs/hub/models-libraries) (such as Transformers or Diffusers). This integration significantly increases your model's accessibility and provides users with code snippets for working with your model. | ||
| |  |  | |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure how this will render in a table, given that the left button is very wide. I tried to check here, but images are not shown for some reason:
| | Code snippet tab | | ||
|
|
||
| Finally, you can adopt the [Mixin class](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) when pushing custom PyTorch models. | ||
| You can also [create your own model library](https://huggingface.co/docs/hub/en/models-adding-libraries) or add Hub support to your library and codebase so they know how to download model weights from the Hub. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| You can also [create your own model library](https://huggingface.co/docs/hub/en/models-adding-libraries) or add Hub support to your library and codebase so they know how to download model weights from the Hub. | |
| You can also [register your own model library](https://huggingface.co/docs/hub/en/models-adding-libraries) or add Hub support to your library and codebase so they know how to download model weights from the Hub. |
To note: I have added an
assetsfolder, which will be deleted later, once I have a green signal on the assets. I will add the assets to the hub, and link the URL.The PR is made to refactor the current model checklist, with an idea of being more concise and actionable insights. I felt the doc also did not have example images, so have added them.
I am open to feedback as this has not been reviewed by and HF team member in the past, so I am open to closing it as well if there is no need of this at the current moment. 🤗