fsdp2_dev

ascend-robot[Feature]opensoraplan1.3新增动态DPCP切换功能

8da061b7创建于 2025年11月6日历史提交

文件	最后提交记录	最后更新时间
data_dynamic_resolution.json	!1054 【Refactor】Modify the configuration in the data.json; T2VDataset uses the factory class to create videoprocessor. Merge pull request !1054 from zs-Derrick/dataset_rebuild_video_processor	1 年前
data_static_resolution.json	!1054 【Refactor】Modify the configuration in the data.json; T2VDataset uses the factory class to create videoprocessor. Merge pull request !1054 from zs-Derrick/dataset_rebuild_video_processor	1 年前
eval_t2v.sh	!1029 [Refactor]configuration file reconstruction Merge pull request !1029 from chenpeizhe/master	1 年前
eval_t2v_model.json	!1029 [Refactor]configuration file reconstruction Merge pull request !1029 from chenpeizhe/master	1 年前
inference_t2v.sh	!1029 [Refactor]configuration file reconstruction Merge pull request !1029 from chenpeizhe/master	1 年前
inference_t2v_model.json	!1029 [Refactor]configuration file reconstruction Merge pull request !1029 from chenpeizhe/master	1 年前
pretrain_t2v.sh	!1443 [Bugfix] fix ckptformat & fix incorrect permissions for ckpt files Merge pull request !1443 from zhangxubin/master	10 个月前
pretrain_t2v_model.json	[Feature]opensoraplan1.3新增动态DPCP切换功能 Co-authored-by: qusongyun1<qusongyun1@noreply.gitcode.com> # message auto-generated for no-merge-commit merge: !1677 merge dynamicDPCP into master [Feature]opensoraplan1.3新增动态DPCP切换功能 Created-by: qusongyun1 Commit-by: qusongyun1 Merged-by: ascend-robot Description: ## Motivation 当前静态DPCP方案在动态负载下无法充分利用算力，例如在大量短序列和少量长序列的情况下，为了保证不OOM，需要设置较大的CP，然而短序列进行大CP并行会导致性能的下降。本特性新增动态DPCP功能，支持在每轮训练迭代中根据数据特征动态切换DP/CP并行策略。 ## Modification pretrain_sora.py：如果开启了动态DPCP，则优先获取缓存数据 training.py: 在初始化时，新增DPCP并行组的初始化，切换后，将数据在cp组内广播并放入缓存 MindSpeed-MM/mindspeed_mm/utils 中新增dpcp_utils.py文件，所有本特性相关的函数实现均在该文件中 ## Self-test (Optional) If modifications to this PR may cause/fix function/accuracy/performance DTSs/issues, a self-inspection record needs to be attached. ## BC-breaking (Optional) If there are compatibility issues, such as dependencies on cann/torch_npu versions, they need to be explained in the PR. ## Checklist **Before PR**: - [x] The new code needs to comply with the Clean Code specification. - [x] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [x] CLA has been signed and all committers have signed the CLA in this PR. - [x] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!1677	7 个月前
samples_prompt.txt	!326 【bugfix】修改readme & 修复tp cp混合时broadcast timestep错误 Merge pull request !326 from zhushixiang/zsx-plan13-1118	1 年前