Add safe-serialization to FullModelHFCheckpointer

[jeffrey-fong]

Context

What is the purpose of this PR? Is it to

• add a new feature
• fix a bug
• update tests and/or documentation
• other (please add here)

Please link to any issues this PR addresses.

As mentioned in #931, FullModelHFCheckpointer saves model checkpoints as .pt files, which is not entirely compatible with HuggingFace Transformers, specifically in Text-Generation Inference (TGI). As a step towards making this repo more compatible with HuggingFace libraries, I propose this PR which allows saving of model checkpoint directly as safetensors.

Changelog

In this PR, I added the option to save model checkpoints as safetensors checkpoint for FullModelHFCheckpointer. This will be done when the option safe-serialization is set to True. The option will default to False if it is set specifically or not set at all.

Example Config for checkpointer:

checkpointer:
  _component_: torchtune.utils.FullModelHFCheckpointer
  checkpoint_dir: /tmp/Meta-Llama-3-8B-Instruct/
  checkpoint_files: [
    model-00001-of-00004.safetensors,
    model-00002-of-00004.safetensors,
    model-00003-of-00004.safetensors,
    model-00004-of-00004.safetensors
  ]
  recipe_checkpoint: null
  output_dir: /tmp/Meta-Llama-3-8B-Instruct/
  model_type: LLAMA3
  safe_serialization: True
resume_from_checkpoint: False

Test plan

Please make sure to do each of the following if applicable to your PR. (If you're not sure about any one of these just ask and we will happily help.)

• run pre-commit hooks and linters (make sure you've first installed via pre-commit install)
• add unit tests for any new functionality
• update docstrings for any new or updated methods or classes
• run unit tests via pytest tests
• run recipe tests via pytest tests -m integration_test
• manually run any new or modified recipes with sufficient proof of correctness
  • include relevant commands and any other artifacts in this summary (pastes of loss curves, eval results, etc.)

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchtune/1096

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 745573e with merge base ef6e196 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[facebook-github-bot]

Hi @jeffrey-fong!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

[facebook-github-bot]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

[joecummings]

This is awesome, thanks @jeffrey-fong! Waiting for tests to pass, then I'll merge.

[joecummings]

🚀

[joecummings]

Looks like this was opened before #1095 was merged. Can you rebase?

[jeffrey-fong]

Sure it's done