Skip to content

Add split argument to required builders and set it default value to "train" #1783

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 2 commits into from
Oct 9, 2024
Merged

Add split argument to required builders and set it default value to "train" #1783

merged 2 commits into from
Oct 9, 2024

Conversation

krammnic
Copy link
Collaborator

@krammnic krammnic commented Oct 9, 2024 •
edited
Loading

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.
#1770

Changelog

What are the changes made in this PR?

  • Add split argument to required builders and set it default value to "train"

Test plan

Please make sure to do each of the following if applicable to your PR. If you're unsure about any one of these just ask and we will happily help. We also have a contributing page for some guidance on contributing.

  • 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.)

UX

If your function changed a public API, please add a dummy example of what the user experience will look like when calling it.
Here is a docstring example
and a tutorial example

  • I did not change any public API
  • I have added an example to docs or docstrings

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Oct 9, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchtune/1783

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

✅ No Failures

As of commit 0b7ea86 with merge base 209d55d (image):
💚 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
Copy link

Hi @krammnic!

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!

@krammnic krammnic changed the title Add split argument to required builders and set it default value to "train Add split argument to required builders and set it default value to "train" Oct 9, 2024
@krammnic
Copy link
Collaborator Author

krammnic commented Oct 9, 2024

@joecummings @RdoubleA Require review

@facebook-github-bot
Copy link

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

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Oct 9, 2024
joecummings
joecummings reviewed Oct 9, 2024
View reviewed changes
Copy link
Member

@joecummings joecummings left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This generally looks good, but can you include some output where you attempt to run on a dataset without specifying train and it starts without any errors?

@krammnic
Copy link
Collaborator Author

krammnic commented Oct 9, 2024 •
edited
Loading

For multimodal:

from torchtune.datasets.multimodal import multimodal_chat_dataset


ds = multimodal_chat_dataset(
    model_transform=None,
    source="Lin-Chen/ShareGPT4V",
    split="train",
    name="ShareGPT4V",
    image_dir="/home/user/dataset/",
    image_tag="<image>",
)

Output:

README.md: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.22k/2.22k [00:00<00:00, 10.9MB/s]
(…)egpt4v_instruct_gpt4-vision_cap100k.json: 100%|██████████████████████████████████████████████████████████████████████████████| 134M/134M [00:02<00:00, 56.2MB/s]
Generating train split: 100%|████████████████████████████████████████████████████████████████████████████████████| 102025/102025 [00:02<00:00, 49778.44 examples/s]

For instruct:

from torchtune.datasets import instruct_dataset

dataset = instruct_dataset(
    tokenizer=None,
    source="json",
    data_files="my_dataset.json",
    column_map={
         "input": "question",
         "output": "answer",
    },
    train_on_input=False,
    packed=False,
)

my_dataset.json

[
    {
        "question": "What time is it in London?",
        "answer": "It is 10:00 AM in London.",
    }
]

Output:

Generating train split: 1 examples [00:00, 122.93 examples/s]

For chat:

from torchtune.datasets import chat_dataset

dataset = chat_dataset(
    tokenizer=None,
    source="json",
    data_files="my_dataset.json",
    conversation_column="conversations",
    conversation_style="sharegpt",
    train_on_input=False,
    packed=False
)

my_dataset.json

[
    {
        "conversations": [
            {
                "from": "human",
                "value": "What time is it in London?",
            },
            {
                "from": "gpt",
                "value": "It is 10:00 AM in London.",
            }
        ]
    }
]

Output:

Generating train split: 1 examples [00:00, 111.08 examples/s]

Should it be covered with test?

@krammnic
Copy link
Collaborator Author

krammnic commented Oct 9, 2024

@joecummings Has provided an examples, but it definitely should be find cause the same argument was passed in **kwargs of SFTDataset

RdoubleA
RdoubleA approved these changes Oct 9, 2024
View reviewed changes
Copy link
Collaborator

@RdoubleA RdoubleA left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is excellent, thank you so much for adding this!

@RdoubleA RdoubleA merged commit a04588e into pytorch:main Oct 9, 2024
17 checks passed
mori360 pushed a commit to mori360/torchtune that referenced this pull request Oct 14, 2024
@krammnic @AsixDO @mori360
Add split argument to required builders and set it default value to "…
ee3e703 
…train" (pytorch#1783)

Co-authored-by: Mark Obozov <obozovmark9@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@joecummings joecummings joecummings left review comments

@RdoubleA RdoubleA RdoubleA approved these changes

Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
5 participants
@krammnic @facebook-github-bot @joecummings @RdoubleA @AsixDO