| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2025-09-18 | 5.1 kB | |
| v0.3.2_ RL Baselines source code.tar.gz | 2025-09-18 | 903.6 kB | |
| v0.3.2_ RL Baselines source code.zip | 2025-09-18 | 969.8 kB | |
| Totals: 3 Items | 1.9 MB | 1 | |
What's Changed
- [misc] set dev version by @hiyouga in https://github.com/hiyouga/EasyR1/pull/372
- fix typo in tensorboard logger by @ning-mz in https://github.com/hiyouga/EasyR1/pull/373
- [utils] fix log probs by @hiyouga in https://github.com/hiyouga/EasyR1/pull/375
- Supports video input by @robinjoe93 in https://github.com/hiyouga/EasyR1/pull/376
- [data] fix video data by @hiyouga in https://github.com/hiyouga/EasyR1/pull/386
- [data] add filter worker by @hiyouga in https://github.com/hiyouga/EasyR1/pull/387
- [example] update qwen3_14b_dapo17k_dapo.sh by @Saigyouji-Yuyuko1000 in https://github.com/hiyouga/EasyR1/pull/389
- [misc] fix ci by @hiyouga in https://github.com/hiyouga/EasyR1/pull/391
- [ckpt] fix remove logic by @hiyouga in https://github.com/hiyouga/EasyR1/pull/393
- [example] update config by @hiyouga in https://github.com/hiyouga/EasyR1/pull/394
- [worker] fix typo in FSDPWorker by @jasper0314-huang in https://github.com/hiyouga/EasyR1/pull/396
- [misc] fix flops counter by @hiyouga in https://github.com/hiyouga/EasyR1/pull/401
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/404
- [readme] add our work to the readme by @JingchengYang4 in https://github.com/hiyouga/EasyR1/pull/408
- [example] change dapo verify by @Saigyouji-Yuyuko1000 in https://github.com/hiyouga/EasyR1/pull/407
- [reward] fix dapo verifier by @hiyouga in https://github.com/hiyouga/EasyR1/pull/410
- Update README.md to include Long-RL by @yukang2017 in https://github.com/hiyouga/EasyR1/pull/411
- [worker] add dynamic batching by @hiyouga in https://github.com/hiyouga/EasyR1/pull/417
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/418
- [worker] fix dp tokens by @hiyouga in https://github.com/hiyouga/EasyR1/pull/419
- [examples] fix config by @hiyouga in https://github.com/hiyouga/EasyR1/pull/420
- [misc] fix ops by @hiyouga in https://github.com/hiyouga/EasyR1/pull/421
- [worker] fix fsdp worker by @hiyouga in https://github.com/hiyouga/EasyR1/pull/422
- [worker] fix grad norm by @hiyouga in https://github.com/hiyouga/EasyR1/pull/423
- [data] better mm data collate by @hiyouga in https://github.com/hiyouga/EasyR1/pull/424
- [trainer] support auto resume by @hiyouga in https://github.com/hiyouga/EasyR1/pull/425
- [worker] add dynamic batching computational workload balance by @hiyouga in https://github.com/hiyouga/EasyR1/pull/426
- [readme] update usage of apptainer by @yzoaim in https://github.com/hiyouga/EasyR1/pull/434
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/442
- [protocol] non blocking false by default by @hiyouga in https://github.com/hiyouga/EasyR1/pull/445
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/447
- [feat] support ray.timeline by @yzoaim in https://github.com/hiyouga/EasyR1/pull/449
- [docker] upgrade vllm to 0.10 by @hiyouga in https://github.com/hiyouga/EasyR1/pull/453
- [worker] fix multi modal data oom by @hiyouga in https://github.com/hiyouga/EasyR1/pull/454
- [misc] fix data proto by @hiyouga in https://github.com/hiyouga/EasyR1/pull/458
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/461
- [trainer] fix checkpoint tracker by @hiyouga in https://github.com/hiyouga/EasyR1/pull/467
- [patch] fix fa utils by @hiyouga in https://github.com/hiyouga/EasyR1/pull/472
- [misc] fix fa patch by @hiyouga in https://github.com/hiyouga/EasyR1/pull/473
- [misc] fix model merger by @hiyouga in https://github.com/hiyouga/EasyR1/pull/479
- [misc] lint by @hiyouga in https://github.com/hiyouga/EasyR1/pull/480
- Fix valset loading for videos by @zhuohaoyu in https://github.com/hiyouga/EasyR1/pull/482
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/486
- [bugfix] fix position ids for latest transformers by @hiyouga in https://github.com/hiyouga/EasyR1/pull/494
- [readme] update wechat by @hiyouga in https://github.com/hiyouga/EasyR1/pull/495
- [misc] pin transformers to 4.56.1 by @hiyouga in https://github.com/hiyouga/EasyR1/pull/496
- [deps] upgrade transformers to 4.54 by @hiyouga in https://github.com/hiyouga/EasyR1/pull/501
- [release] v0.3.2 by @hiyouga in https://github.com/hiyouga/EasyR1/pull/502
New Contributors
- @ning-mz made their first contribution in https://github.com/hiyouga/EasyR1/pull/373
- @robinjoe93 made their first contribution in https://github.com/hiyouga/EasyR1/pull/376
- @jasper0314-huang made their first contribution in https://github.com/hiyouga/EasyR1/pull/396
- @JingchengYang4 made their first contribution in https://github.com/hiyouga/EasyR1/pull/408
- @yukang2017 made their first contribution in https://github.com/hiyouga/EasyR1/pull/411
- @yzoaim made their first contribution in https://github.com/hiyouga/EasyR1/pull/434
- @zhuohaoyu made their first contribution in https://github.com/hiyouga/EasyR1/pull/482
Full Changelog: https://github.com/hiyouga/EasyR1/compare/v0.3.1...v0.3.2