0
代码
代码
Issues
Pull Requests
讨论
Wiki
分析
0
ascend-robotascend-robot[Docs]Add documentation related to the loss calculation method of VLM Model
ddd5a373创建于 2025年10月25日历史提交
文件最后提交记录最后更新时间
__init__.py
!1321 [Bugfix] import dependency isolation. Merge pull request !1321 from zs-Derrick/master 10 个月前
adaptive_clip_grad_patch.py
!1417 [Feature] wan2.1 add FSDP2 config, and patch clip grad. Merge pull request !1417 from zs-Derrick/master 9 个月前
canonical_layer_patch.py
[Feature] Support canonical layer training in videoalign Co-authored-by: chenpeizhe<chenpeizhe1@huawei.com> # message auto-generated for no-merge-commit merge: merge master into master [Feature] Support canonical layer training in videoalign Created-by: chenpeizhe Commit-by: chenpeizhe Merged-by: ascend-robot Description: ## Motivation Support qwen2vl canonical model (same as transformers) in videoalign training. ## Modification Support qwen2vl canonical model (same as transformers) in videoalign training for accuracy mode. ## 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!15118 个月前
dummy_optimizer_patch.py
!1360 [Bugfix] remove redundancy patch for dummy optimizer in 012 Merge pull request !1360 from chenhaihui/dummy_optimizer_bugfix 9 个月前
fsdp1_patches.py
!1384 [Feature] support lumina-mgpt2 model Merge pull request !1384 from meng-coding/master 9 个月前
infer_fa_patch.py
!1312 [Refactor] Add inference fa patch Merge pull request !1312 from 王泽/infer_fa_patch 10 个月前
meta_device_patches.py
[feature] qwen2.5vl support fsdp2 Co-authored-by: cxiaolong<2845907121@qq.com> # message auto-generated for no-merge-commit merge: merge master into master [feature] qwen2.5vl support fsdp2 Created-by: cxiaolong Commit-by: cxiaolong 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**: - [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!15258 个月前
models_patches.py
!1391 [Refactor] Remove Unused Patches: SequentialMLP_forward, adamw, diffusers_geglu, t5 Merge pull request !1391 from zhangxubin/master 9 个月前
patch_manager.py
!1491 [Feature] Add loading .pt checkpoint file when using distributed ckpt format Merge pull request !1491 from zs-Derrick/master 8 个月前
ring_attn_patch.py
!1382 [Refactor] Adapt latest mindspeed Merge pull request !1382 from 王泽/latest012 8 个月前
training_patches.py
!1491 [Feature] Add loading .pt checkpoint file when using distributed ckpt format Merge pull request !1491 from zs-Derrick/master 8 个月前
ulysses_patches.py
!1382 [Refactor] Adapt latest mindspeed Merge pull request !1382 from 王泽/latest012 8 个月前
validate_args_patch.py
[Docs]Add documentation related to the loss calculation method of VLM Model Co-authored-by: zhangxubin<1656631289@qq.com> # message auto-generated for no-merge-commit merge: !1642 merge 2.2.0 into 2.2.0 [Docs]Add documentation related to the loss calculation method of VLM Model Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## Motivation Add documentation related to the loss calculation method of VLM Model. ## Modification Add documentation related to the loss calculation method of VLM Model. ## 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!16427 个月前