【Bugfix】Correct the spelling errors in the code
Co-authored-by: lu-jinfu1999<lujinfu1@h-partners.com>
# message auto-generated for no-merge-commit merge:
!1670 merge master into master
【Bugfix】Correct the spelling errors in the code
Created-by: lu-jinfu1999
Commit-by: lu-jinfu1999
Merged-by: ascend-robot
Description: ## Motivation
Correct the spelling errors in the code.
## Modification
Correct spelling errors in the document.
Correct spelling errors in the code.
## 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!1670
[mindspore] support qwen3vl finetune for megatron backend
Co-authored-by: ffmh<fengminghao2@huawei.com>
# message auto-generated for no-merge-commit merge:
!1785 merge qwen3vl_megatron into master
[mindspore] support qwen3vl finetune for megatron backend
Created-by: ffmh
Commit-by: ffmh
Merged-by: ascend-robot
Description: ## Motivation
Please describe the motivation of this PR and the goal you want to achieve through this PR.
适配 qwen3vl 8B/30B megatron后端
## Modification
Please briefly describe what modification is made in this PR.
添加 layer_spec修改、text模块transformer block/layer添加deepstack相关适配
## 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!1785
【Bugfix】Correct the spelling errors in the code
Co-authored-by: lu-jinfu1999<lujinfu1@h-partners.com>
# message auto-generated for no-merge-commit merge:
!1670 merge master into master
【Bugfix】Correct the spelling errors in the code
Created-by: lu-jinfu1999
Commit-by: lu-jinfu1999
Merged-by: ascend-robot
Description: ## Motivation
Correct the spelling errors in the code.
## Modification
Correct spelling errors in the document.
Correct spelling errors in the code.
## 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!1670
[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
[Feature]VACE Model and Data Process
Co-authored-by: feng0w0<houyufeng4@huawei.com>
# message auto-generated for no-merge-commit merge:
!1669 merge master into master
[Feature]VACE Model and Data Process
Created-by: feng0w0
Commit-by: feng0w0
Merged-by: ascend-robot
Description: ## Motivation
Support VACE Model
## Modification
Added VACE data processing pipeline and model components.
## 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!1669
[Test] chunkloss UT
Co-authored-by: liyingxuan<liyingxuan3@huawei.com>
# message auto-generated for no-merge-commit merge:
!1751 merge chunkloss_ut into master
[Test] chunkloss UT
Created-by: liyx616
Commit-by: liyingxuan
Merged-by: ascend-robot
Description: ## Motivation
chunkloss ut
## Modification
增加chunkloss ut，并修复两个笔误
## 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!1751
[feature] qwen2.5vl 72b support vit-dp
Co-authored-by: cxiaolong<2845907121@qq.com>
# message auto-generated for no-merge-commit merge:
!1565 merge master into master
[feature] qwen2.5vl 72b support vit-dp
Created-by: cxiaolong
Commit-by: cxiaolong
Merged-by: ascend-robot
Description: ## Motivation
1. For long-sequence scenarios, the QwenVL series models currently already support RingAttention/Ulysses/Hybrid CP in ViT.
2. vit-dp is a new long-sequence solution suitable for video understanding tasks, designed to split the sequences of ViT modules.
3. vit-dp use communications free approach to split ViT sequence, compared to Ulysses and RingAttention CP, achieving better performance compared to Ulysses.
## 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!1565
[Bugfix]resolve multiple issues — unused code, index out of bounds, undefined vars, resource leaks
Co-authored-by: zhangxubin<1656631289@qq.com>
# message auto-generated for no-merge-commit merge:
!1662 merge master into master
[Bugfix]resolve multiple issues — unused code, index out of bounds, undefined vars, resource leaks
Created-by: MoCuishle-M
Commit-by: zhangxubin
Merged-by: ascend-robot
Description: ## Motivation
Fix some security issues.
## Modification
The issues fixed are as follows:
1. Removed unused code and fixed logic errors
2. Fixed array out-of-bounds access.
3. Fixed usage of undefined variables
4. Fixed resource leaks by ensuring proper release
## 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!1662
[Feature] qwen3vl improve performance
Co-authored-by: liyingxuan<liyingxuan3@huawei.com>
# message auto-generated for no-merge-commit merge:
!1717 merge master into master
[Feature] qwen3vl improve performance
Created-by: liyx616
Commit-by: liyingxuan
Merged-by: ascend-robot
Description: ## Motivation
qwen3vl improve performance
## Modification
1. moe block融合算子（默认关闭）
2. rmsnorm和rope融合算子
## 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!1717
[mindspore] support qwen3vl finetune for megatron backend
Co-authored-by: ffmh<fengminghao2@huawei.com>
# message auto-generated for no-merge-commit merge:
!1785 merge qwen3vl_megatron into master
[mindspore] support qwen3vl finetune for megatron backend
Created-by: ffmh
Commit-by: ffmh
Merged-by: ascend-robot
Description: ## Motivation
Please describe the motivation of this PR and the goal you want to achieve through this PR.
适配 qwen3vl 8B/30B megatron后端
## Modification
Please briefly describe what modification is made in this PR.
添加 layer_spec修改、text模块transformer block/layer添加deepstack相关适配
## 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!1785
[Bugfix]resolve multiple issues — unused code, index out of bounds, undefined vars, resource leaks
Co-authored-by: zhangxubin<1656631289@qq.com>
# message auto-generated for no-merge-commit merge:
!1662 merge master into master
[Bugfix]resolve multiple issues — unused code, index out of bounds, undefined vars, resource leaks
Created-by: MoCuishle-M
Commit-by: zhangxubin
Merged-by: ascend-robot
Description: ## Motivation
Fix some security issues.
## Modification
The issues fixed are as follows:
1. Removed unused code and fixed logic errors
2. Fixed array out-of-bounds access.
3. Fixed usage of undefined variables
4. Fixed resource leaks by ensuring proper release
## 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!1662
