| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
[feature] support ep resume Co-authored-by: htwang<wanghaitao60@huawei.com> # message auto-generated for no-merge-commit merge: !2143 merge fsdp2_dev into fsdp2_dev [feature] support ep resume Created-by: htwang Commit-by: htwang Merged-by: ascend-robot Description: ## Motivation 支持EP场景的断点续训 ## Modification 1、修改EP实现方式（使用distribute module实现） 2、优化器状态适配EP场景保存和加载 ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2143 | 4 个月前 | |
[Feature] MM_FSDP2: Add train_epochs to control the number of iterations. Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2215 merge fsdp2_dev into fsdp2_dev [Feature] MM_FSDP2: Add train_epochs to control the number of iterations. Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation MM_FSDP2: Add train_epochs to control the number of iterations. ## Modification 新增 train_epochs参数，其指定时，与len(data_loader)，计算得到train_iters ## 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!2215 | 4 个月前 | |
[Feature] qwen3.5 support expert parallel Co-authored-by: young256<liumingyang16@huawei.com> # message auto-generated for no-merge-commit merge: !2210 merge fsdp2_dev_qwen35_publish into fsdp2_dev [Feature] qwen3.5 support expert parallel Created-by: young256 Commit-by: young256 Merged-by: ascend-robot Description: ## Motivation Qwen3.5 MoE requires enabling expert parallelism at the 397B parameter scale to prevent Out‑of‑Memory (OOM) errors. ## Modification Supporting expert parallel for Qwen3.5 MoE ## Self-test (Optional) not involve yet ## BC-breaking (Optional) not involve ## 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!2210 | 4 个月前 | |
[feature] recompute support use_reentrant Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2216 merge fsdp2_dev into fsdp2_dev [feature] recompute support use_reentrant Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2216 | 4 个月前 | |
[feature] Apply chunkloss lm head abstraction to feature and keep raw loss func Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2172 merge fsdp2_dev into fsdp2_dev [feature] Apply chunkloss lm head abstraction to feature and keep raw loss func Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2172 | 4 个月前 | |
[Feature] MM_FSDP2: Add train_epochs to control the number of iterations. Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2215 merge fsdp2_dev into fsdp2_dev [Feature] MM_FSDP2: Add train_epochs to control the number of iterations. Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation MM_FSDP2: Add train_epochs to control the number of iterations. ## Modification 新增 train_epochs参数，其指定时，与len(data_loader)，计算得到train_iters ## 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!2215 | 4 个月前 | |
[Feature]MM_FSDP2: add fused npu operator. Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2162 merge fsdp2_dev4 into fsdp2_dev [Feature]MM_FSDP2: add fused npu operator. Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation MM_FSDP2: add fused npu operator. ## Modification add fused npu operator. ## 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!2162 | 4 个月前 | |
[Feature]MM_FSDP2: ModelHub support both Transformers models and Custom models Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2149 merge fsdp2_dev into fsdp2_dev [Feature]MM_FSDP2: ModelHub support both Transformers models and Custom models Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation MM_FSDP2: ModelHub support both Transformers models and Custom models ## Modification MM_FSDP2: ModelHub support both Transformers models and Custom models ## 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!2149 | 4 个月前 | |
[Feature] keep profiling of fsdp2 dev the same with the master branch Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2196 merge fsdp2_dev into fsdp2_dev [Feature] keep profiling of fsdp2 dev the same with the master branch Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation keep profiling same with the master branch ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2196 | 4 个月前 | |
[Feature] Support CosyVoice3 finetuning Co-authored-by: AZe_404<wangze62@h-partners.com> # message auto-generated for no-merge-commit merge: !2183 merge cosyvoice3_train into fsdp2_dev [Feature] Support CosyVoice3 finetuning Created-by: AZe_404 Commit-by: AZe_404 Merged-by: ascend-robot Description: ## Motivation 支持CosyVoice3模型llm微调. ## Modification 1. 新增模型yaml配置文件及微调启动脚本； 2. 适配最新 Trainer，接入自定义dataloader 3. 支持DataSetArguments及DataloaderArguments参数类to_dict()函数包含未定义额外参数 ## 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!2183 | 4 个月前 | |
[Feature] keep profiling of fsdp2 dev the same with the master branch Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2196 merge fsdp2_dev into fsdp2_dev [Feature] keep profiling of fsdp2 dev the same with the master branch Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation keep profiling same with the master branch ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2196 | 4 个月前 | |
[Feature] keep profiling of fsdp2 dev the same with the master branch Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2196 merge fsdp2_dev into fsdp2_dev [Feature] keep profiling of fsdp2 dev the same with the master branch Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation keep profiling same with the master branch ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2196 | 4 个月前 | |
[Feature] keep profiling of fsdp2 dev the same with the master branch Co-authored-by: weixin_44031810<gaojie75@huawei.com> # message auto-generated for no-merge-commit merge: !2196 merge fsdp2_dev into fsdp2_dev [Feature] keep profiling of fsdp2 dev the same with the master branch Created-by: gaojie_ Commit-by: weixin_44031810 Merged-by: ascend-robot Description: ## Motivation keep profiling same with the master branch ## Modification Please briefly describe what modification is made in this PR. ## 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**: - [ ] The new code needs to comply with the Clean Code specification. - [ ] The PR content is self-checked, and the expression can be clear and the writing standardized **After PR**: - [ ] CLA has been signed and all committers have signed the CLA in this PR. - [ ] The ci-pipeline is passed, Code Check is passed. See merge request: Ascend/MindSpeed-MM!2196 | 4 个月前 | |
[Feature] MM_FSDP2: Add attention patch for transformers. And add ring & usp for qwen3vl Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2120 merge fsdp2_dev2 into fsdp2_dev [Feature] MM_FSDP2: Add attention patch for transformers. And add ring & usp for qwen3vl Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation Add attention patch for transformers. And add ring & usp for qwen3vl ## Modification 1.Add transformers attention patch to support cp（ulysses/ring/usp) 2.add ring&usp for qwen3vl ## 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!2120 | 4 个月前 |