0
代码
代码
Issues
Pull Requests
讨论
Wiki
分析
0
ascend-robotascend-robot[docs] modify qwen2.5vl mindspeed branch to support A5 training
af6e3684创建于 3月10日历史提交
文件最后提交记录最后更新时间
common
feat(torch): Squash merge fsdp2_dev into master. Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2223 merge master into master feat(torch): Squash merge fsdp2_dev into master. Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation Squash merge fsdp2_dev into master. ## Modification fsdp2_dev分支合并至master ## 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!22233 个月前
fsdp
feat(torch): Squash merge fsdp2_dev into master. Co-authored-by: zs-derrick1<1434012475@qq.com> # message auto-generated for no-merge-commit merge: !2223 merge master into master feat(torch): Squash merge fsdp2_dev into master. Created-by: zs-derrick1 Commit-by: zs-derrick1 Merged-by: ascend-robot Description: ## Motivation Squash merge fsdp2_dev into master. ## Modification fsdp2_dev分支合并至master ## 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!22233 个月前
sora_model
feat(torch): fsdp2 wan2.2 14b t2v support lora finetune Co-authored-by: LKONE<wanglikai4@huawei.com> # message auto-generated for no-merge-commit merge: !2218 merge master into master feat(torch): fsdp2 wan2.2 14b t2v support lora finetune Created-by: wanglikai1019 Commit-by: LKONE 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!22183 个月前
vlm_model
[Feature] qwen3.5 support new dcp convert and fused rmsnorm Co-authored-by: young256<liumingyang16@huawei.com> # message auto-generated for no-merge-commit merge: !2246 merge fsdp2_dev_qwen35_publish into master [Feature] qwen3.5 support new dcp convert and fused rmsnorm Created-by: young256 Commit-by: young256 Merged-by: ascend-robot Description: ## What this PR does / why we need it? Qwen3.5 needs to support new DCP weight conversion to reduce the vector operator overhead caused by weight transpose in the program. The mul vector operator caused by RMSNorm and RMSNormGate needs to be eliminated. ## Does this PR introduce any user-facing change? DCP weight conversion instructions have changed. ## How was this patch tested? The weight verification after weight conversion is valid. The fused operator replacement has completed precision alignment. See merge request: Ascend/MindSpeed-MM!22462 个月前
__init__.py
[docs] modify qwen2.5vl mindspeed branch to support A5 training Co-authored-by: LKONE<wanglikai4@huawei.com> # message auto-generated for no-merge-commit merge: !2262 merge master into master [docs] modify qwen2.5vl mindspeed branch to support A5 training Created-by: wanglikai1019 Commit-by: LKONE Merged-by: ascend-robot Description: ## What this PR does / why we need it? modify qwen2.5vl mindspeed branch to support A5 training ## Does this PR introduce any user-facing change? modify qwen2.5vl mindspeed branch to support A5 training ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-MM!22622 个月前
convert_cli.py
[feature]support moe expert merge weight convert from hf to dcp or vice versa Co-authored-by: pjgao<gaopengju3@huawei.com> # message auto-generated for no-merge-commit merge: !1752 merge master into master [feature]support moe expert merge weight convert from hf to dcp or vice versa Created-by: PIPIXIU Commit-by: pjgao Merged-by: ascend-robot Description: ## Motivation InternVL3.5使用MOE专家合并的方式需要权重转换，当前仓上是两段转换，先将原仓safetensors转换成合并专家后的safetensors，再转换成torch dcp，因此本PR新增转换接口mm-convert ExpertMergeDcpConverter hf_to_dcp /mm-convert ExpertMergeDcpConverter dcp_to_hf ## Modification 1. 增加转换类ExpertMergeDcpConverter 2. 抽象Commandable管理所有的命令行接口，提供add_command接口，允许增加函数到命令行中 ## 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**: - [ ] 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!17526 个月前