Skip to content
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

making gpt2 fx traceable #34633

Merged
merged 3 commits into from
Nov 25, 2024

Conversation

xuzifei-dmatrix
Copy link
Contributor

What does this PR do?

This PR fixes the issue that gpt2 is no longer fx traceable in the latest version
Below code gives error complaining about control flows not traceable. This PR solves the issue.

from transformers.utils.fx import HFTracer
from transformers import pipeline
import torch
from torch.fx import GraphModule

pipe = pipeline(
    task="text-generation",
    model="openai-community/gpt2",
    device="cuda",
)
x = torch.randint(1, 100, (1, 1024)).to("cuda")
past_key_values = pipe.model(x, use_cache=True).past_key_values
dummy_inputs = dict(input_ids=x, labels=x, past_key_values=past_key_values)

tracer = HFTracer()
gm = GraphModule(pipe.model, tracer.trace(pipe.model, dummy_inputs=dummy_inputs))

Fixes # (issue)

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you read the contributor guideline,
    Pull Request section?
  • Was this discussed/approved via a Github issue or the forum? Please add a link
    to it if that's the case.
  • Did you make sure to update the documentation with your changes? Here are the
    documentation guidelines, and
    here are tips on formatting docstrings.
  • Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@ArthurZucker

Copy link
Collaborator

@ArthurZucker ArthurZucker left a comment

Choose a reason for hiding this comment

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

Thanks, do you mind running make fix-copies?

Copy link
Collaborator

@ArthurZucker ArthurZucker left a comment

Choose a reason for hiding this comment

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

LGTM thanks!

@ArthurZucker
Copy link
Collaborator

This indeed fixes the issue of:

TraceError: Proxy object cannot be iterated. This can be attempted when the Proxy is used in a loop or as a *args or **kwargs function argument. See the torch.fx docs on pytorch.org for a more detailed explanation of what types of control flow can be traced, and check out the Proxy docstring for help troubleshooting Proxy iteration errors

@ArthurZucker ArthurZucker merged commit bfc3556 into huggingface:main Nov 25, 2024
26 checks passed
@HuggingFaceDocBuilderDev

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.

BernardZach pushed a commit to BernardZach/transformers that referenced this pull request Dec 5, 2024
* making gpt2 fx tracable

* running make fix-copies

* Revert "running make fix-copies"

This reverts commit 5a3437c.
BernardZach pushed a commit to innovationcore/transformers that referenced this pull request Dec 6, 2024
* making gpt2 fx tracable

* running make fix-copies

* Revert "running make fix-copies"

This reverts commit 5a3437c.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants