Skip to content

Conversation

@chenyushuo
Copy link
Collaborator

Description

As the title says.

Checklist

Please check the following items before code is ready to be reviewed.

  • Code has passed all tests
  • Docstrings have been added/updated in Google Style
  • Documentation has been updated
  • Code is ready for review

@chenyushuo chenyushuo requested a review from pan-x-c December 31, 2025 05:22
@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @chenyushuo, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical bug in the multi-stage resume functionality by ensuring that all stage configurations, including those for already completed stages, are properly updated. This prevents potential inconsistencies and ensures correct behavior when resuming training. The changes also include an enhanced test case to validate the system's error handling for invalid configurations and improve test isolation by using deep copies of configuration objects.

Highlights

  • Bug Fix: Multi-Stage Resume Configuration: Resolved an issue where stage_config.check_and_update() was not being called for completed stages during a multi-stage resume, potentially leading to outdated configurations for skipped stages.
  • Enhanced Error Handling Test: Added a new test case to test_trainer that explicitly verifies the system raises an exception when an invalid taskset path is provided in a multi-stage configuration, improving the robustness of error detection.
  • Improved Test Isolation: Implemented deepcopy when handling configuration objects within tests to ensure that test modifications do not inadvertently affect the original configuration, thereby enhancing test reliability and preventing side effects.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

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

Code Review

This pull request addresses a bug in the multi-stage resume functionality. The core fix in trinity/cli/launcher.py correctly ensures that configurations for skipped stages are updated, which is crucial for subsequent stages that depend on them. The changes in tests/trainer/trainer_test.py provide a solid test case for this scenario, simulating a failed stage and a successful resume. The use of deepcopy to prevent config mutation within the test is a great improvement for test isolation and reliability. I have one suggestion to make the exception assertion in the test more specific.

@chenyushuo
Copy link
Collaborator Author

/unittest-module-trainer

@github-actions
Copy link

Summary

Tests 📝 Passed ✅ Failed ❌ Skipped ⏭️ Other ❓ Flaky 🍂 Duration ⏱️
24 21 0 3 0 0 35m 27s

Skipped

Tests Status
tests/trainer/trainer_test.py::TestMultiModalGRPO::test_trainer skipped ⏭️
tests/trainer/trainer_test.py::TestMultiModalSFT::test_trainer skipped ⏭️
tests/trainer/trainer_test.py::TestTinkerTrainer::test_trainer skipped ⏭️

Tests

Test Name Status Flaky Duration
tests/trainer/trainer_test.py::TestTrainerCountdown_0_fsdp::test_trainer 3m 26s
tests/trainer/trainer_test.py::TestTrainerCountdown_1_megatron::test_trainer 3m 37s
tests/trainer/trainer_test.py::TestStepAheadAsyncRL::test_trainer 1m 46s
tests/trainer/trainer_test.py::TestTrainerGSM8K_0_fsdp::test_trainer 51.3s
tests/trainer/trainer_test.py::TestTrainerGSM8K_1_fsdp2::test_trainer 1m 16s
tests/trainer/trainer_test.py::TestTrainerGSM8K_2_fsdp::test_trainer 51.8s
tests/trainer/trainer_test.py::TestTrainerGSM8K_3_fsdp2::test_trainer 59.2s
tests/trainer/trainer_test.py::TestTrainerSFTWarmupGSM8K::test_trainer 2m 2s
tests/trainer/trainer_test.py::TestTrainerDPO::test_trainer 37.1s
tests/trainer/trainer_test.py::TestTrainerSFT::test_trainer 34.2s
tests/trainer/trainer_test.py::TestTrainerToolsSFT::test_trainer_tools 35.5s
tests/trainer/trainer_test.py::TestFullyAsyncMode_0_fsdp::test_fully_async_mode 1m 34s
tests/trainer/trainer_test.py::TestFullyAsyncMode_1_fsdp::test_fully_async_mode 1m 32s
tests/trainer/trainer_test.py::TestFullyAsyncMode_2_megatron::test_fully_async_mode 2m 17s
tests/trainer/trainer_test.py::TestTrainerCheckpointSave_0_fsdp::test_trainer 1m 42s
tests/trainer/trainer_test.py::TestTrainerCheckpointSave_1_megatron::test_trainer 3m 29s
tests/trainer/trainer_test.py::TestTrainerMIX::test_trainer 1m 31s
tests/trainer/trainer_test.py::TestServeWithTrainer::test_serve_with_trainer 1m 49s
tests/trainer/trainer_test.py::TestMultiModalGRPO::test_trainer ⏭️ 781ms
tests/trainer/trainer_test.py::TestMultiModalSFT::test_trainer ⏭️ 531ms
tests/trainer/trainer_test.py::TestTrainerLoRA::test_trainer 2m 45s
tests/trainer/trainer_test.py::TestOverRollout::test_trainer 46.4s
tests/trainer/trainer_test.py::TestTrainerPromptTruncation::test_trainer 1m 11s
tests/trainer/trainer_test.py::TestTinkerTrainer::test_trainer ⏭️ 537ms

Github Test Reporter by CTRF 💚

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.

1 participant