Refactor data module and support task pipeline in data processor #92

HYLcool · 2025-06-19T09:46:34Z

Description

Refactor the data module according to the latest buffer implementations
Add RawDataReader for reading data from raw dataset files
Support task pipeline in the data processor part and add corresponding examples
Update docs about data examples
Fix several bugs

TODO in future PRs:

Support experience pipeline.

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

+ add new raw data reader and writer buffer * support raw data reader and writer in RftDataset - remove db exporting in RftDataset and ActiveIterator

# Conflicts: # tests/buffer/file_test.py # trinity/buffer/reader/file_reader.py # trinity/buffer/writer/file_writer.py

* fix: make dirs when initializing file writer

# Conflicts: # tests/buffer/file_test.py # trinity/buffer/writer/file_writer.py

docs/sphinx_doc/source/tutorial/example_data_functionalities.md

trinity/cli/launcher.py

# Conflicts: # docs/sphinx_doc/source/tutorial/example_data_functionalities.md # trinity/buffer/buffer.py # trinity/buffer/reader/file_reader.py # trinity/cli/launcher.py

# Conflicts: # trinity/cli/launcher.py

pan-x-c · 2025-06-20T08:45:16Z

/run-unittest

github-actions · 2025-06-20T09:04:21Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Pending ⏳	Other ❓	Flaky 🍂	Duration ⏱️
38	38	0	0	0	0	0	1.1s

Failed Tests

No failed tests ✨

Flaky Tests

No flaky tests ✨

Skipped

No skipped tests ✨

Tests

Test Name	Status	Duration
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_dpo_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_mix_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_opmd_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_ppo_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_sft_policy_loss	✅	1ms
tests/buffer/file_test.py::TestFileBuffer::test_file_buffer	✅	3ms
tests/buffer/file_test.py::TestFileBuffer::test_file_reader	✅	1ms
tests/buffer/file_test.py::TestFileBuffer::test_file_writer	✅	2ms
tests/buffer/queue_test.py::TestQueueBuffer::test_queue_buffer	✅	2ms
tests/buffer/sql_test.py::TestSQLBuffer::test_create_sql_buffer	✅	3ms
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	1ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	5ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/vllm_test.py::TestModelWrapperSyncV0::test_generate	✅	41ms
tests/common/vllm_test.py::TestModelWrapperAsyncV0::test_generate	✅	40ms
tests/common/vllm_test.py::TestModelWrapperAsyncTPV0::test_generate	✅	50ms
tests/common/vllm_test.py::TestModelWrapperAsyncTPV1::test_generate	✅	50ms
tests/common/vllm_test.py::TestModelWrapperAsyncV1::test_generate	✅	38ms
tests/common/vllm_test.py::TestAPIServer::test_api	✅	24ms
tests/common/vllm_test.py::TestTokenizer::test_assistant_token_mask	✅	1ms
tests/explorer/explorer_test.py::BaseExplorerCase::test_explorer	✅	1ms
tests/explorer/explorer_test.py::TestExplorerCountdownEval::test_explorer	✅	98ms
tests/explorer/explorer_test.py::TestExplorerCountdownNoEval::test_explorer	✅	129ms
tests/explorer/runner_pool_test.py::RunnerPoolTest::test_runner_pool	✅	22ms
tests/explorer/runner_pool_test.py::RunnerPoolTest::test_runner_pool_with_auxiliary_models	✅	4ms
tests/explorer/workflow_test.py::WorkflowTest::test_gsm8k_workflow	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_complex_workflow	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_fraction_workflow	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_workflow	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_workflow_resettable	✅	1ms
tests/trainer/trainer_test.py::BaseTrainerCase::test_trainer	✅	1ms
tests/trainer/trainer_test.py::TestTrainerCountdown::test_trainer	✅	276ms
tests/trainer/trainer_test.py::TestStepAheadAsyncRL::test_trainer	✅	105ms
tests/trainer/trainer_test.py::TestTrainerGSM8K::test_trainer	✅	73ms
tests/trainer/trainer_test.py::TestTrainerSFTWarmupGSM8K::test_trainer	✅	69ms
tests/trainer/trainer_test.py::TestTrainerDPO::test_trainer	✅	53ms
tests/utils/plugin_test.py::TestPluginLoader::test_load_plugins	✅	4ms

Github Test Reporter by CTRF 💚

yxdyc

LGTM

HYLcool added 23 commits June 13, 2025 12:06

* reformat existing configs

e0874db

+ add new raw data reader and writer buffer * support raw data reader and writer in RftDataset - remove db exporting in RftDataset and ActiveIterator

* after pre-commit

a1ce540

* modified according to the latest buffer imp.

a63b5f4

* sort by priority before writing to buffer

2c7fd58

* refine args and messages

d60a931

+ Add tests for file buffer

086fade

Merge branch 'main' into refactor/data

e81ae48

# Conflicts: # tests/buffer/file_test.py # trinity/buffer/reader/file_reader.py # trinity/buffer/writer/file_writer.py

* merge from main

4ab6676

* after pre-commit

ec45e8c

* fix task parser

c15ef5f

* update .gitignore

2bdcdea

* update .gitignore

7511c96

* fix wrong way to check the types of processors in task parser

c93ff10

* fix: make dirs when initializing file writer

* update data dependencies

66eaece

Merge branch 'main' into refactor/data

40ee1b6

# Conflicts: # tests/buffer/file_test.py # trinity/buffer/writer/file_writer.py

* create dir first before writing to files

55480e3

* create dir first before writing to files

a888035

+ add missing finish method for file_wrapper

ec7f7c6

+ add trust_remote_code=True to all load_dataset invoking

d73f89c

* only consider numeric stats

df5e4db

* make task pipeline work

6163ec5

* modify docs for task pipeline

253f2e1

* after pre-commit

ef781fd

HYLcool requested review from chenyushuo, pan-x-c, yanxi-chen and yxdyc June 19, 2025 09:46

HYLcool self-assigned this Jun 19, 2025

HYLcool added the enhancement New feature or request label Jun 19, 2025

* rename unit test

1666099

pan-x-c reviewed Jun 20, 2025

View reviewed changes

docs/sphinx_doc/source/tutorial/example_data_functionalities.md Outdated Show resolved Hide resolved

docs/sphinx_doc/source/tutorial/example_data_functionalities.md Outdated Show resolved Hide resolved

trinity/cli/launcher.py Outdated Show resolved Hide resolved

* modify according to xuchen's comments

2baed2e

pan-x-c changed the base branch from main to algorithm_dev June 20, 2025 08:15

HYLcool and others added 3 commits June 20, 2025 16:25

Merge branch 'algorithm_dev' into refactor/data

8384b9c

# Conflicts: # docs/sphinx_doc/source/tutorial/example_data_functionalities.md # trinity/buffer/buffer.py # trinity/buffer/reader/file_reader.py # trinity/cli/launcher.py

Merge branch 'algorithm_dev' into refactor/data

9e2653b

# Conflicts: # trinity/cli/launcher.py

fix example config

4fb41a1

yxdyc approved these changes Jun 20, 2025

View reviewed changes

pan-x-c merged commit eddf4e4 into algorithm_dev Jun 20, 2025
3 checks passed

HYLcool deleted the refactor/data branch June 20, 2025 09:49

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Refactor data module and support task pipeline in data processor #92

Refactor data module and support task pipeline in data processor #92

Uh oh!

HYLcool commented Jun 19, 2025 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pan-x-c commented Jun 20, 2025

Uh oh!

github-actions bot commented Jun 20, 2025

Uh oh!

yxdyc left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Refactor data module and support task pipeline in data processor #92

Refactor data module and support task pipeline in data processor #92

Uh oh!

Conversation

HYLcool commented Jun 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pan-x-c commented Jun 20, 2025

Uh oh!

github-actions bot commented Jun 20, 2025

Summary

Failed Tests

Flaky Tests

Skipped

Tests

Uh oh!

yxdyc left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

HYLcool commented Jun 19, 2025 •

edited

Loading