Batch Size for RL training
Nobody has claimed this yet.
Assessment
- Difficulty
- 1/5
- Estimated time
- Under an hour
- Newbie friendliness
- 25/100
- Issue type
- Documentation
- Clarity
- Needs clarification
- Activity status
- Stale
- Tech stack
- machine-learning
- Domain
- machine-learning
Research direction
Review the report's description of RL training and locate the definitions of batch size, rollout number, and mini-batches. Confirm whether the reported batch size includes rollouts and state how many mini-batches are used during training; the issue is done when both questions have a clear answer.
Written by the indexing model from the issue text.
Description
Thanks for this great work!
In the report, it's mentioned that a batch size of 128 is used. Is this including the rollout number or not?
Also, how many mini-batches are used during training?
Thanks
- Dominant language
- No language data
- Stars
- 759
- Forks
- 59
- PR merge metrics
- No merged PRs in 30d
Getting set up
We have not checked this project's setup files yet. Start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from ByteDance-Seed/Seed-Coder
-
Difficulty 2/5 1-3 hours Newbie friendliness 45/100
ByteDance-Seed/Seed-Coder#24 ·
-
Difficulty 5/5 Over a week Newbie friendliness 20/100
ByteDance-Seed/Seed-Coder#23 ·
-
Difficulty 5/5 Over a week Newbie friendliness 25/100
ByteDance-Seed/Seed-Coder#22 ·
-
Difficulty 5/5 Over a week Newbie friendliness 25/100
ByteDance-Seed/Seed-Coder#21 ·
-
Difficulty 5/5 Over a week Newbie friendliness 20/100
ByteDance-Seed/Seed-Coder#20 ·
All issues in ByteDance-Seed/Seed-Coder
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
jingyaogong/minimind#880 ·
Maintainers usually reply within 1 day
-
Qwen3_5MoeModel no longer returns router_logits, breaking aux loss with output_router_logits=TrueOpen
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
huggingface/transformers#49172 ·
Maintainers usually reply within 1 day
-
[Bug]: MRV2 prepare_inputs signature is incompatible with updated vLLM num_active_loras argumentOpenbug
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
vllm-project/vllm-ascend#17710 ·
Maintainers usually reply within 1 day
-
bug good first issue
Difficulty 2/5 1-3 hours Newbie friendliness 90/100
selimfirat/pysad#209 ·
Maintainers usually reply within 1 day
-
backend::vllm diffusion multimodal
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 1 day