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

switch to grpc for deploy and eval #11643

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
Open

Conversation

HuiyingLi
Copy link
Collaborator

@HuiyingLi HuiyingLi commented Dec 18, 2024

What does this PR do ?

Using PyTriton grpc client instead of rest API when running evaluation due to the large logits payload.
On arc_challenge with bs=1, llama3-8b
Before (rest API): 9487.118s
After (gRPC client): 502.976

Collection: [Note which collection this PR will affect]

Changelog

  • Add specific line by line info of high level changes in this PR.

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this 

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

  • New Feature
  • Bugfix
  • Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

  • Related to # (issue)

HuiyingLi and others added 3 commits December 17, 2024 20:05
Copy link
Collaborator

@hemildesai hemildesai left a comment

Choose a reason for hiding this comment

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

Great speedup, I am wondering would it be possible to remove the rest option all together from eval? There doesn't seem to be any benefit in keeping the eval with rest API as an option.

@HuiyingLi
Copy link
Collaborator Author

Great speedup, I am wondering would it be possible to remove the rest option all together from eval? There doesn't seem to be any benefit in keeping the eval with rest API as an option.

Thanks for the suggestion! Updated the PR and removed the rest API part.

Copy link
Contributor

[🤖]: Hi @HuiyingLi 👋,

We wanted to let you know that a CICD pipeline for this PR just finished successfully

So it might be time to merge this PR or get some approvals

I'm just a bot so I'll leave it you what to do next.

//cc @pablo-garay @ko3n1g

return np.char.encode(str_ndarray, "utf-8")


def query_llm(
Copy link
Collaborator

Choose a reason for hiding this comment

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

@HuiyingLi can we reuse the query_llm from here so that we can avoid redundant code ?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

@athitten Thank you for pointing out! Just changed to use the exisiting query_llm.

Copy link
Contributor

beep boop 🤖: 🙏 The following files have warnings. In case you are familiar with these, please try helping us to improve the code base.


Your code was analyzed with PyLint. The following annotations have been identified:

************* Module nemo.collections.llm.api
nemo/collections/llm/api.py:442:0: C0301: Line too long (130/119) (line-too-long)
nemo/collections/llm/api.py:568:0: C0116: Missing function or method docstring (missing-function-docstring)
nemo/collections/llm/api.py:15:0: W0611: Unused import os (unused-import)
************* Module nemo.deploy.deploy_base
nemo/deploy/deploy_base.py:37:0: C0115: Missing class docstring (missing-class-docstring)
nemo/deploy/deploy_base.py:71:4: C0116: Missing function or method docstring (missing-function-docstring)
nemo/deploy/deploy_base.py:75:4: C0116: Missing function or method docstring (missing-function-docstring)
nemo/deploy/deploy_base.py:79:4: C0116: Missing function or method docstring (missing-function-docstring)
nemo/deploy/deploy_base.py:83:4: C0116: Missing function or method docstring (missing-function-docstring)
nemo/deploy/deploy_base.py:114:4: C0116: Missing function or method docstring (missing-function-docstring)
************* Module nemo.deploy.deploy_pytriton
nemo/deploy/deploy_pytriton.py:53:0: C0301: Line too long (132/119) (line-too-long)
nemo/deploy/deploy_pytriton.py:84:0: C0301: Line too long (126/119) (line-too-long)

-----------------------------------
Your code has been rated at 9.65/10

Mitigation guide:

  • Add sensible and useful docstrings to functions and methods
  • For trivial methods like getter/setters, consider adding # pylint: disable=C0116 inside the function itself
  • To disable multiple functions/methods at once, put a # pylint: disable=C0116 before the first and a # pylint: enable=C0116 after the last.

By applying these rules, we reduce the occurance of this message in future.

Thank you for improving NeMo's documentation!

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

Successfully merging this pull request may close these issues.

3 participants