Support conversation_style of openai format (OpenAI API style)

[xingyaoww]

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.

This adds a conversation_style of openai, similar to the openai one by @musabgultekin in #824 (comment).

Changelog

What are the changes made in this PR?

• Add conversion function of openai_to_llama2_messages

It now supports converting sample like (following OpenAI API) the following directly to internal llama2_messages.

{
    # key could be "messages" OR "conversations"
    "messages": [
        {
            "role": <system|user|assistant>,
            "content": <message>,
        },
        ...
    ]
}

Example usage in config:

# Dataset
dataset:
  _component_: torchtune.datasets.chat_dataset
  source: json
  data_files: data/datasets/codeact-mixture.jsonl
  split: train
  conversation_style: standard_chat
  chat_format: ChatMLFormat
  max_seq_len: 16384
seed: 42
shuffle: True

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/890

• 📄 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 95f2386 with merge base ea3d4ea ():
💚 Looks good so far! There are no failures yet. 💚

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

[RdoubleA]

Thanks @xingyaoww for this addition! Overall looks great, just some minor comments.

Would be great if you can add a unit test for this as well. I also realized the sharegpt converter is missing a unit test also (I know, how could we) but both of these should be tested. Upto you if you also want to cover that :)

[xingyaoww]

@RdoubleA Thanks a ton for the review!!! I should fixed all the test issues in the latest commit and can confirm all tests passed!

[RdoubleA]

this looks good to go! and thanks for the additional test! let me start the github CI and see what it thinks

[RdoubleA]

remove this line?

[xingyaoww]

done!

[RdoubleA]

@xingyaoww Looks like just a lint error which you should be able to fix with pre-commit

[xingyaoww]

@RdoubleA should be fixed!