Skip to content

Add example for experience replay #345

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
chenyushuo merged 4 commits into from
Oct 27, 2025
Merged

Add example for experience replay #345

chenyushuo merged 4 commits into from
Oct 27, 2025

Conversation

@yanxi-chen
Copy link
Collaborator

@yanxi-chen yanxi-chen commented Oct 27, 2025

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

@gemini-code-assist
Copy link
Contributor

gemini-code-assist bot commented Oct 27, 2025

Summary of Changes

Hello @yanxi-chen, 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 adds a comprehensive example for implementing experience replay with Proximal Policy Optimization (PPO) on the Countdown dataset within the Trinity-RFT system. The primary goal is to demonstrate how experience replay can enhance learning efficiency and optimize hardware utilization by reusing rollout samples, particularly in scenarios where agent-environment interaction is resource-intensive. The example provides a concrete configuration and explains the custom priority function used, supported by experimental evidence of its positive impact.

Highlights

  • New Example for Experience Replay: This pull request introduces a new example demonstrating the experience replay mechanism within the Trinity-RFT framework, specifically for PPO on the Countdown dataset.
  • Detailed Configuration: A new countdown.yaml configuration file is added, showcasing how to enable and configure experience replay, including settings for the buffer, reuse_cooldown_time, and a custom priority_fn.
  • Custom Priority Function: The example utilizes a linear_decay_use_count_control_randomization priority function, whose logic is detailed to prioritize fresher and less used samples, with an option for random perturbation.
  • Experimental Results: The accompanying README.md presents experimental results that compare experience replay against a baseline, illustrating its benefits in learning efficiency and the ability to take more training steps.
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. ↩

gemini-code-assist[bot]
gemini-code-assist bot reviewed Oct 27, 2025
View reviewed changes
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 introduces an example of experience replay using the PPO algorithm on the Countdown dataset. It includes a README.md explaining the implementation and configuration, and a countdown.yaml file with the specific configurations for this example. I have provided comments to address potential issues related to clarity and completeness.

chenyushuo
chenyushuo reviewed Oct 27, 2025
View reviewed changes
@pan-x-c
Copy link
Collaborator

pan-x-c commented Oct 27, 2025

/unittest-module-common

@github-actions
Copy link

github-actions bot commented Oct 27, 2025

Summary

Tests 📝 Passed ✅ Failed ❌ Skipped ⏭️ Other ❓ Flaky 🍂 Duration ⏱️
33 33 0 0 0 0 317ms

Tests

Test Name Status Flaky Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid ✅ 32ms
tests/common/config_test.py::TestConfig::test_config_flatten ✅ 1ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid ✅ 1ms
tests/common/config_test.py::TestConfig::test_default_workflow ✅ 1ms
tests/common/config_test.py::TestConfig::test_load_default_config ✅ 3ms
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly ✅ 1ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster ✅ 2ms
tests/common/experience_test.py::TestEID::test_eid_properties ✅ 1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type ✅ 1ms
tests/common/experience_test.py::TestExperience::test_assertions ✅ 1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience ✅ 1ms
tests/common/experience_test.py::TestExperience::test_gather ✅ 1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion ✅ 1ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience ✅ 1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize ✅ 1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience ✅ 1ms
tests/common/experience_test.py::TestExperience::test_to_dict ✅ 1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion ✅ 1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion ✅ 1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion ✅ 1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields ✅ 1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion ✅ 1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate ✅ 57ms
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate ✅ 36ms
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate ✅ 45ms
tests/common/vllm_test.py::TestModelLen_0::test_model_len ✅ 21ms
tests/common/vllm_test.py::TestModelLen_1::test_model_len ✅ 21ms
tests/common/vllm_test.py::TestAPIServer::test_api ✅ 24ms
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async ✅ 24ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask ✅ 1ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools ✅ 1ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls ✅ 22ms
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls ✅ 20ms

Github Test Reporter by CTRF 💚

@chenyushuo chenyushuo merged commit 15c296f into modelscope:main Oct 27, 2025
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@chenyushuo chenyushuo chenyushuo approved these changes

+1 more reviewer

@gemini-code-assist gemini-code-assist[bot] gemini-code-assist[bot] left review comments

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@yanxi-chen @pan-x-c @chenyushuo