| feat: add skill - mindspeed-mm-fsdp2-model-only-migration Co-authored-by: hanyyy<hanyue42@huawei.com> # message auto-generated for no-merge-commit merge: !2595 merge model-migration into master feat: add skill - mindspeed-mm-fsdp2-model-only-migration Created-by: vasileone Commit-by: hanyyy Merged-by: ascend-robot Description: ## What this PR does / why we need it? 新增 mindspeed-mm-fsdp2-model-only-migration skill，用于指导仅模型开源、缺少训练 pipeline 的模型迁移到 MindSpeed-MM 插件式 FSDP2 后端。 同时移除旧的 VLM 专用 skill，补充 FSDP2 框架地图 reference 和迁移流程示意图，提升模型迁移流程的泛化性和可执行性。 ## Does this PR introduce any user-facing change? 无 ## How was this patch tested? 文档/skill 更新，不涉及运行测试。 See merge request: Ascend/MindSpeed-MM!2595 | 3 个月前 |
| feat: Update Docker image overview & refactor image build process Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2975 merge master into master feat: Update Docker image overview & refactor image build process Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. 重构镜像构建过程 1.1 改变dockerfile写法，使得镜像大小从25.9GB 缩减至 14.9GB。 通过指令顺序与缓存优化，多阶段构建（最终镜像只保留运行时必需的产物）实现镜像瘦身。 1.2 删除一些不使用的接口 1.3 剥离CI镜像构建逻辑 1.4 删除qwen3vl GRPO环境构建。该模型已经通过verl仓库承载。 2. 调整CI镜像构建参数，为后续的版本升级做准备 3. 根据用户反馈与相关要求更新overview文档，提供镜像易用性。 ## Does this PR introduce any user-facing change? 更改了镜像构建使用方法，已经在overview里更新说明。 ## How was this patch tested? 本地验证镜像构建没问题。 验证了三种chip、两种OS。 See merge request: Ascend/MindSpeed-MM!2975 | 1 个月前 |
| docs: modify pull request template Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2225 merge master into master docs: modify pull request template Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? 按照最新的标准修改pr模板 ## Does this PR introduce any user-facing change? 不涉及代码功能 ## How was this patch tested? 不涉及 See merge request: Ascend/MindSpeed-MM!2225 | 6 个月前 |
| docs:rename Ascend Extension for PyTorch Co-authored-by: z60112595<zhaotao68@h-partners.com> # message auto-generated for no-merge-commit merge: !2836 merge master into master docs:rename Ascend Extension for PyTorch Created-by: kali00 Commit-by: z60112595 Merged-by: ascend-robot Description: ## What this PR does / why we need it? https://gitcode.com/Ascend/MindSpeed-MM/issues/506 1、更新Ascend Extension for PyTorch名称为TorchNPU 2、补充快速入门环境最低配置 ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2836 | 2 个月前 |
| 修复 qwen25vl_7B mbridge 特性报错的问题 Co-authored-by: bonjour_rzq1998<ruanzeqi1@huawei.com> # message auto-generated for no-merge-commit merge: !2906 merge feature/fix-mbridge-qwen25vl into master 修复 qwen25vl_7B mbridge 特性报错的问题 Created-by: bonjour_rzq1998 Commit-by: bonjour_rzq1998 Merged-by: ascend-robot Description: What this PR does / why we need it? 1、修改bridge/models/qwen_vl/__init__.py中对于异常情况的容忍范围。 Does this PR introduce any user-facing change? 不涉及 How was this patch tested? 不涉及 See merge request: Ascend/MindSpeed-MM!2906 | 1 个月前 |
| feat: Adapt for save_async feature Co-authored-by: LKONE<wanglikai4@huawei.com> # message auto-generated for no-merge-commit merge: !2825 merge MM-0712 into master feat: Adapt for save_async feature Created-by: wanglikai1019 Commit-by: LKONE Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. 校正dcp_to_hf函数的缩进语法问题，从八个空格改为四个空格； 2. 修改save_async配置说明，当前仅支持dcp权重异步保存，非dcp权重保存且同时开启了异步保存功能时，打印提示； 3. save_checkpointer.save方法调用时传入save_async参数。 ## Does this PR introduce any user-facing change? 无 ## How was this patch tested? 测试dcp权重异步保存功能是否正常。 See merge request: Ascend/MindSpeed-MM!2825 | 2 个月前 |
| feat: Update Docker image overview & refactor image build process Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2975 merge master into master feat: Update Docker image overview & refactor image build process Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. 重构镜像构建过程 1.1 改变dockerfile写法，使得镜像大小从25.9GB 缩减至 14.9GB。 通过指令顺序与缓存优化，多阶段构建（最终镜像只保留运行时必需的产物）实现镜像瘦身。 1.2 删除一些不使用的接口 1.3 剥离CI镜像构建逻辑 1.4 删除qwen3vl GRPO环境构建。该模型已经通过verl仓库承载。 2. 调整CI镜像构建参数，为后续的版本升级做准备 3. 根据用户反馈与相关要求更新overview文档，提供镜像易用性。 ## Does this PR introduce any user-facing change? 更改了镜像构建使用方法，已经在overview里更新说明。 ## How was this patch tested? 本地验证镜像构建没问题。 验证了三种chip、两种OS。 See merge request: Ascend/MindSpeed-MM!2975 | 1 个月前 |
| docs: Update OVERVIEW.md and OVERVIEW.zh.md Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2980 merge master into master docs: Update OVERVIEW.md and OVERVIEW.zh.md Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? supported_tags.md 链接换成绝对路径（URL） 因为 OVERVIEW.md、 OVERVIEW.zh.md 需要在ascendhub 与 quay.io 呈现。 如果写成相对路径，ascendhub、quay.io上进行链接跳转时，因为文件不存在，所导致404 如果第一次提交PR时，写成URL，但PR还没合入，该URL不存在 会导致文档检测过不了。 所以先使用相对路径合入，再使用第二个PR改成 URL形式。 ## Does this PR introduce any user-facing change? / ## How was this patch tested? / See merge request: Ascend/MindSpeed-MM!2980 | 1 个月前 |
| docs: fix text pretraining configuration guide Co-authored-by: yeqm<yeqiangmao@huawei.com> # message auto-generated for no-merge-commit merge: !2974 merge docs/fix-text-pretrain-guide into master docs: fix text pretraining configuration guide Created-by: ye_qm Commit-by: yeqm Merged-by: ascend-robot Description: ## What this PR does / why we need it? The FSDP2 text pretraining example in docs/zh/features/pretrain.md does not match the current implementation. The existing example only sets attr.pretrain: true. However, the FSDP2 backend selects PretrainDatasetProcessor through basic_parameters.stage. Without stage: pretrain, the stage defaults to sft, causing text-only samples without a response field to be dropped and eventually resulting in StopIteration. This PR: - Adds the required basic_parameters.stage: pretrain setting for the FSDP2 backend. - Replaces the deprecated collate_param.model_name with collate_param.collator_id. - Clarifies the configuration differences between the FSDP2 and Megatron backends. - Corrects the packing description: text is packed within each preprocessing batch, so each batch must contain at least cutoff_len valid tokens to generate a training sample. - Improves the formatting and readability of the guide. Related issue: N/A. The problem was found while verifying the documented workflow against the current master implementation. ## Does this PR introduce any user-facing change? Yes. This PR corrects the user-facing configuration required to run text-only pretraining with the FSDP2 backend. Users should configure: ```yaml data: dataset_param: attr: formatting: alpaca prompt: text basic_parameters: stage: pretrain template: default dataloader_param: collate_param: collator_id: llm_pretrain See merge request: Ascend/MindSpeed-MM!2974 | 1 个月前 |
| [docs] merge kimi-k2.7-code examples into kimi-k2.5, keep version differences in README only Co-authored-by: LKONE<wanglikai4@huawei.com> # message auto-generated for no-merge-commit merge: !2969 merge MM-KimiK2.7 into master [docs] merge kimi-k2.7-code examples into kimi-k2.5, keep version differences in README only Created-by: wanglikai1019 Commit-by: LKONE Merged-by: ascend-robot Description: ## What this PR does / why we need it? （1）刷新Kimi-K2.6的文档； （2）刷新Kimi相关特性的文档。 ## Does this PR introduce any user-facing change? 无 ## How was this patch tested? 无 See merge request: Ascend/MindSpeed-MM!2969 | 1 个月前 |
| fix(data): accept positional state_dict in PrefetchGradAccDataLoader.load_state_dict Co-authored-by: iceflysnow<iceflysnow@126.com> # message auto-generated for no-merge-commit merge: !2949 merge fix/prefetch-grad-acc-load-state-dict-positional into master fix(data): accept positional state_dict in PrefetchGradAccDataLoader.load_state_dict Created-by: iceflysnow Commit-by: iceflysnow Merged-by: ascend-robot Description: ## 关联 Issue #578 — https://gitcode.com/Ascend/MindSpeed-MM/issues/578 ## 问题概述 loss_cfg.loss_type: per_token_loss 时，训练侧用 PrefetchGradAccDataLoader 包裹基础 dataloader，其 load_state_dict 为 keyword-only（**kwargs），而 TrainEngine.load() 在 resume 时以位置参数调用它，二者不匹配，断点续训时全 rank 崩溃 TypeError: ...takes 1 positional argument but 2 were given，退出码 1，无法从 checkpoint 恢复训练。 激活条件：PrefetchGradAccDataLoader 仅在 loss_type == "per_token_loss" 时启用（trainer.py L383-385）；default loss_type 下用 torchdata 库的基础 StatefulDataLoader（其 load_state_dict(state_dict) 接受位置参数，调用不冲突），故不崩——属合法常见配置（Intern-S2 等官方示例在用）下的真实缺陷。详见 issue。 ## 修复方案 让 PrefetchGradAccDataLoader.load_state_dict 接受位置参数，与调用方（train_engine.py:378 现状）及被包裹的基础 StatefulDataLoader（均按位置传参）签名对齐。该签名同时兼容位置调用与原 keyword 调用，不引入行为变化。 mindspeed_mm/fsdp/data/dataloader/dataloader.py： diff @@ -249,7 +249,11 @@ class PrefetchGradAccDataLoader: def state_dict(self): return self.base_dataloader.state_dict() - def load_state_dict(self, **kwargs): - self.base_dataloader.load_state_dict(**kwargs) + def load_state_dict(self, state_dict=None): + # Accept a positional state_dict (the train_engine caller passes it + # positionally). Forward to the wrapped StatefulDataLoader which also + # takes a single positional state_dict argument. + if state_dict is not None: + self.base_dataloader.load_state_dict(state_dict) ## 验证 环境：8 × 910B2，CANN 25.3.RC1，torch_npu 2.7.1，模型 Qwen3.5-9B（LoRA r=8）。 - **基线复现（修复前）**：per_token_loss 配置 run1 存全量 DCP（含 optim/rng），run2 resume 时 Loaded checkpoint 后到 dataloader 恢复处即崩：TypeError: PrefetchGradAccDataLoader.load_state_dict() takes 1 positional argument but 2 were given（train_engine.py:378），退出码 1，全 rank 崩。 - **回归（修复后）**：run2 干净 resume：Loaded checkpoint → iteration 4/5 | consumed samples: 64 → iteration 5/5 | consumed samples: 80，rc=0，无 TypeError；iteration 与 consumed_train_samples 与 run1 末尾连续。 - **长测稳定性**：run1 60 iter 全量 save@20/40/60（rc=0），隐藏 iter60 让 run2 从 iter40 resume → iter41–60，consumed 672 → 960 连续，loss 与 run1 末段衔接，全程零 TypeError/Traceback/Error，rc=0。 > 验证在官方 master 852ddffd 上完成；最新 master 81c9cd7d（本 PR base）未触及 dataloader.py（两版本该文件逐字节相同），故结论等价迁移。 See merge request: Ascend/MindSpeed-MM!2949 | 1 个月前 |
| feat: add gitleaks secret detection Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2817 merge master into master feat: add gitleaks secret detection Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. 引入gitleaks二进制离线扫描工具 2. 新增pre-commit/.gitleaks.toml配置，继承官方全部检测规则 3. 配置pre-commit钩子，提交前自动扫描密钥硬编码风险。 ## Does this PR introduce any user-facing change? 无 ## How was this patch tested? PR流水线pre-commit检测新增敏感信息检测. See merge request: Ascend/MindSpeed-MM!2817 | 2 个月前 |
| feat: for the FSDP2 backend README, always pull from the MindSpeed master branch. Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2864 merge master into master feat: for the FSDP2 backend README, always pull from the MindSpeed master branch. Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 使用 fsdp2 后端的模型，修改成每次都拉取最新的MindSpeed master 分支代码 详细变更： 1. scripts/install.sh 新增分支参数支持 - 新增 -b, --msbranch BRANCH 参数，用于指定 MindSpeed 分支名 - 与 --msid 互斥，同时指定时报错退出 - 克隆时使用 git clone -b <branch> 一步完成分支切换；已存在目录时使用 git fetch origin <branch> && git checkout <branch> 保持最新 - 配置打印与安装日志根据参数显示 branch/commit/最新版 2. fsdp2 后端 README 统一改为拉取 master 分支 - 判断逻辑：脚本中存在 torchrun $DISTRIBUTED_ARGS mindspeed_mm/fsdp/train/trainer.py 即代表使用 fsdp2 后端 - 使用 bash scripts/install.sh 的 README 增加 --msbranch master 参数 - 使用手动 git clone 的 README 改为 git clone -b master - 存在 README_v1.md 的目录（qwen3vl、qwen3omni）只修改 v1，不动 README.md 涉及文件： - scripts/install.sh 、 examples/qwen3_5/README.md （首次提交） - examples/kimik2_5/README.md 、 examples/ltx2/README.md 、 examples/mimov2_5/README.md 、 examples/moss_tts/README.md 、 examples/qwen3_6/README.md 、 examples/qwen3omni/README_v1.md 、 examples/qwen3tts/README.md 、 examples/qwen3vl/README_v1.md 、 examples/step3_vl/README.md （第二次提交） ## Does this PR introduce any user-facing change? 是。 使用 fsdp2 后端的用户安装流程变化： - 原方式： bash scripts/install.sh --msid eb10b92 或手动 git clone https://gitcode.com/Ascend/MindSpeed.git - 新方式： bash scripts/install.sh --msbranch master 或 git clone -b master https://gitcode.com/Ascend/MindSpeed.git 用户不再需要查询和维护 MindSpeed commit ID，每次安装自动拉取最新 master 分支代码。 相关文档： - scripts/install.sh - 新增 -b/--msbranch 参数（ --msid 保留向后兼容） - examples/qwen3_5/README.md - examples/qwen3_6/README.md - examples/mimov2_5/README.md - examples/kimik2_5/README.md - examples/step3_vl/README.md - examples/qwen3tts/README.md - examples/moss_tts/README.md - examples/ltx2/README.md - examples/qwen3omni/README_v1.md - examples/qwen3vl/README_v1.md ## How was this patch tested? 1. 语法校验： 通过 bash -n scripts/install.sh 验证脚语法正确 2. 互斥校验： 同时指定 --msid 和 --msbranch 时正确报错退出 3. 参数解析： 验证 -b master 和 --msbranch master 两种写法均可正确解析 4. 克隆逻辑： 新目录场景使用 git clone -b ，已存在目录场景使用 fetch + checkout 使用约束与限制： - --msid 参数保留向后兼容，未修改的 README（bagel、hunyuanvideo_1.5、wan2.2、qwen3vl/README.md、mcore/qwen3_5 等）仍使用 commit ID 方式 - 分支名必须为远端存在的有效分支，否则 git fetch / git clone -b 会失败 - 离线环境（无法访问 gitcode.com）不适用此方式，仍需使用 --msid 配合本地缓存 See merge request: Ascend/MindSpeed-MM!2864 | 2 个月前 |
| docs: update wechat qrcode Co-authored-by: bonjour_rzq1998<ruanzeqi1@huawei.com> # message auto-generated for no-merge-commit merge: !2961 merge feature/MM-wechat-qrcode-change into master docs: update wechat qrcode Created-by: bonjour_rzq1998 Commit-by: bonjour_rzq1998 Merged-by: ascend-robot Description: What this PR does / why we need it? Update sources/images/MM_wechat_qrcode.jpg with the latest QR code for the MindSpeed MM community group. Does this PR introduce any user-facing change? Yes. Users can scan the updated QR code in the README to join the latest community group. How was this patch tested? The image was resized to 757 × 757 pixels to match the existing format. The processed QR code was successfully decoded and verified to be scannable. See merge request: Ascend/MindSpeed-MM!2961 | 1 个月前 |
| feat(fsdp): add autograd anomaly detection Co-authored-by: ye_qm<yeqiangmao@huawei.com> # message auto-generated for no-merge-commit merge: !2966 merge yqm/autograd-anomaly-detection into master feat(fsdp): add autograd anomaly detection Created-by: ye_qm Commit-by: ye_qm Merged-by: ascend-robot Description: ``markdown ## What this PR does / why we need it? PyTorch autograd errors, such as backward operator failures and NaN gradients, can be difficult to locate because the error may not identify the corresponding forward operation. This PR adds environment-variable-controlled autograd anomaly detection for the MindSpeed-MM FSDP2 training path. Detailed changes: - Register a new FSDP2 environment variable: - Name: MM_DETECT_ANOMALY - Type: bool - Default: false - Read the environment variable through the centralized FSDP2 environment-variable manager. - Configure PyTorch anomaly detection during trainer initialization: python detect_anomaly = envs.MM_DETECT_ANOMALY torch.autograd.set_detect_anomaly(detect_anomaly) - Enable anomaly detection before model execution so backward failures can include the corresponding forward traceback. - Print a warning when anomaly detection is enabled because it significantly reduces training performance. - Apply the setting through the common Trainer.initialize() path, covering both the standard FSDP2 trainer and the FunASR trainer. - Add tests for: - the default disabled state; - boolean environment-variable conversion; - propagation of the environment-variable value to PyTorch. - Add troubleshooting documentation describing how to enable the feature and its performance limitations. - Replace the direct LOCAL_RANK environment read in model_container_hub.py with envs.LOCAL_RANK, keeping FSDP2 business code consistent with centralized environment-variable management. This is a debugging feature rather than a bugfix, so there is no related issue. ## Does this PR introduce any user-facing change? Yes. Users can enable autograd anomaly detection through the following environment variable: bash export MM_DETECT_ANOMALY=true The environment variable supports the project-wide boolean formats: - True values: 1, true, yes, y, on - False values: 0, false, no, n, off Values are case-insensitive and may contain surrounding whitespace. If MM_DETECT_ANOMALY is unset, it defaults to false. Therefore, existing training scripts and default runtime behavior are unchanged. To disable it explicitly: bash export MM_DETECT_ANOMALY=false Related documentation: text docs/zh/troubleshooting.md ## How was this patch tested? The following checks were performed: - Environment-variable unit tests: bash pytest tests/ut_fsdp/test_envs.py -q Result: text 63 passed - Python syntax compilation for all modified Python files. - git diff --check to detect whitespace errors. - Ruff and Black checks for the modified test files. - GitCode remote Git Hooks checks passed. The following test files cover this feature: text tests/ut_fsdp/test_envs.py tests/ut_fsdp/train/test_autograd_anomaly.py The tests verify that: 1. MM_DETECT_ANOMALY defaults to false when it is not exported. 2. Supported true and false strings are converted to Python bool values. 3. Trainer.initialize() calls: python torch.autograd.set_detect_anomaly(False) when the environment variable is disabled. 4. Trainer.initialize() calls: python torch.autograd.set_detect_anomaly(True) when the environment variable is enabled. 5. FSDP2 business code does not introduce unapproved direct os.getenv() or os.environ.get() calls. The focused trainer test can be run with: bash pytest tests/ut_fsdp/train/test_autograd_anomaly.py -v Usage constraints and limitations: - This feature is intended only for debugging. - Enabling it significantly reduces training performance. - It should remain disabled during formal or performance-sensitive training. - It helps trace backward failures to their corresponding forward operations, but it does not guarantee detection of every numerical issue. - In distributed training, anomaly information is reported by the rank on which the failure occurs. - The local development environment did not contain PyTorch, so the PyTorch-dependent trainer test was added but could not be executed locally. It should be executed by the project CI environment. `` See merge request: Ascend/MindSpeed-MM!2966 | 1 个月前 |
| docs: fix master doc ci Co-authored-by: liuxi_<liuxi75@huawei.com> # message auto-generated for no-merge-commit merge: !2651 merge doc-scan-fix into master docs: fix master doc ci Created-by: liuxi_ Commit-by: liuxi_ Merged-by: ascend-robot Description: ## What this PR does / why we need it? fix doc ci ## Does this PR introduce any user-facing change? fix doc ci ## How was this patch tested? fix doc ci See merge request: Ascend/MindSpeed-MM!2651 | 3 个月前 |
| 修改完善pre-commit工具配置 Co-authored-by: wujinyuan1<wujinyuan1@huawei.com> # message auto-generated for no-merge-commit merge: !2470 merge master into master 修改完善pre-commit工具配置 Created-by: wujinyuan1 Commit-by: wujinyuan1 Merged-by: ascend-robot Description: ## What this PR does / why we need it? pre-commit 工具规则配置修改：试运行阶段发现python bandit工具检测过严，超出昇腾编程规范。 此次修改重点修改bandit规则。 ## Does this PR introduce any user-facing change? 门禁codecheck检测规则精简，提升代码合入效率 ## How was this patch tested? 代码扫描工具配置，不涉及代码仓功能 See merge request: Ascend/MindSpeed-MM!2470 | 4 个月前 |
| feat: Update Docker image overview & refactor image build process Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2975 merge master into master feat: Update Docker image overview & refactor image build process Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. 重构镜像构建过程 1.1 改变dockerfile写法，使得镜像大小从25.9GB 缩减至 14.9GB。 通过指令顺序与缓存优化，多阶段构建（最终镜像只保留运行时必需的产物）实现镜像瘦身。 1.2 删除一些不使用的接口 1.3 剥离CI镜像构建逻辑 1.4 删除qwen3vl GRPO环境构建。该模型已经通过verl仓库承载。 2. 调整CI镜像构建参数，为后续的版本升级做准备 3. 根据用户反馈与相关要求更新overview文档，提供镜像易用性。 ## Does this PR introduce any user-facing change? 更改了镜像构建使用方法，已经在overview里更新说明。 ## How was this patch tested? 本地验证镜像构建没问题。 验证了三种chip、两种OS。 See merge request: Ascend/MindSpeed-MM!2975 | 1 个月前 |
| chore: add VS Code devcontainer configuration. Co-authored-by: zhangxubin<zhangxubin3@h-partners.com> # message auto-generated for no-merge-commit merge: !2911 merge master into master chore: add VS Code devcontainer configuration. Created-by: MoCuishle-M Commit-by: zhangxubin Merged-by: ascend-robot Description: ## What this PR does / why we need it? 当前仓库缺乏 Dev Container 配置（.devcontainer/devcontainer.json 及配套 Dockerfile），导致开发者在不同环境中需要手动配置依赖，降低了协作效率。 为 MindSpeed-MM 新增 VS Code Dev Container 配置（.devcontainer/），一键启动标准化的容器开发环境。 同时对dockerfile做一定适配简化。 ## Does this PR introduce any user-facing change? 可选使用，不影响已有代码。VS Code 打开仓库 → F1 → Reopen in Container 即可。 ## How was this patch tested? 在 arm64 NPU 服务器上 Reopen in Container 验证通过：镜像构建、依赖安装、pre-commit hooks、torch_npu.npu.is_available() 均正常。 See merge request: Ascend/MindSpeed-MM!2911 | 1 个月前 |
| docs: 移除文档中 MindSpore 相关展示 Co-authored-by: liuxi_<liuxi75@huawei.com> # message auto-generated for no-merge-commit merge: !2670 merge docs-fix into master docs: 移除文档中 MindSpore 相关展示 Created-by: liuxi_ Commit-by: liuxi_ Merged-by: ascend-robot Description: ## What this PR does / why we need it? 移除文档中 MindSpore 相关展示(目录树/Profiler项),仅保留 PyTorch 移除出海描述 ## Does this PR introduce any user-facing change? 移除文档中 MindSpore 相关展示(目录树/Profiler项),仅保留 PyTorch 移除出海描述 ## How was this patch tested? read See merge request: Ascend/MindSpeed-MM!2670 | 3 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| !325 【测试】添加InternVL2-8B ST & build打包内容完善 Merge pull request !325 from 陆劲夫/master | 1 年前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| docs: add Kimi-K3 release news | 1 个月前 |
| docs:rename Ascend Extension for PyTorch Co-authored-by: z60112595<zhaotao68@h-partners.com> # message auto-generated for no-merge-commit merge: !2836 merge master into master docs:rename Ascend Extension for PyTorch Created-by: kali00 Commit-by: z60112595 Merged-by: ascend-robot Description: ## What this PR does / why we need it? https://gitcode.com/Ascend/MindSpeed-MM/issues/506 1、更新Ascend Extension for PyTorch名称为TorchNPU 2、补充快速入门环境最低配置 ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2836 | 2 个月前 |
| fix: KimiK3 use fused npu_rms_norm in KimiRMSNorm, fix chunk_kda. Co-authored-by: suis-moi<zhushixiang@huawei.com> # message auto-generated for no-merge-commit merge: !2899 merge master into master fix: KimiK3 use fused npu_rms_norm in KimiRMSNorm, fix chunk_kda. Created-by: suis-moi Commit-by: suis-moi Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1、重构MOE模块 2、更新开源最新版代码 3、修复chunk kda.py 4、替换kimi RMSNORM为融合算子 ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? 本地已经验证通过 See merge request: Ascend/MindSpeed-MM!2899 | 1 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| bugfix: 合法合规风险数据修复 Co-authored-by: yaoyaoxu<xuyaoyao.824404@huawei.com> # message auto-generated for no-merge-commit merge: !2763 merge codecheck_260702 into master bugfix: 合法合规风险数据修复 Created-by: yaoyaoxu Commit-by: yaoyaoxu Merged-by: ascend-robot Description: ## What this PR does / why we need it? 合法合规风险数据修复：补充缺少的Copyright ## Does this PR introduce any user-facing change? 不涉及用户界面修改 ## How was this patch tested? 通过已有ut和st See merge request: Ascend/MindSpeed-MM!2763 | 2 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| [Bugfix] fix protobuf resource conflict bug Co-authored-by: LKONE<wanglikai4@huawei.com> # message auto-generated for no-merge-commit merge: !2343 merge master into master [Bugfix] fix protobuf resource conflict bug Created-by: wanglikai1019 Commit-by: LKONE Merged-by: ascend-robot Description: ## What this PR does / why we need it? 当环境中存在tensorflow时，直接拉起训练脚本，在import transformers时会因框架不兼容导致段错误，程序异常退出 ## Does this PR introduce any user-facing change? 无 ## How was this patch tested? 在装有tensorflow库的环境中拉起训练脚本，看是否能够正常运行 See merge request: Ascend/MindSpeed-MM!2343 | 5 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| [modify] modify the threshold for gc in qwen3_vl 30B Co-authored-by: ye_qm<yeqiangmao@huawei.com> # message auto-generated for no-merge-commit merge: !2593 merge fix-qwen3vl-gc into master [modify] modify the threshold for gc in qwen3_vl 30B Created-by: ye_qm Commit-by: ye_qm Merged-by: ascend-robot Description: ## What this PR does / why we need it? This PR adds an optional GC threshold setting for Qwen3-VL 30B LoRA fine-tuning. Background: During Qwen3-VL 30B training, Python GC behavior may introduce performance fluctuation in some runs. This PR allows the Qwen3-VL 30B LoRA training script to explicitly enable a tuned GC threshold before training starts. Changes: - Adds ENABLE_GC_THRESHOLD=1 to examples/qwen3vl/finetune_lora_qwen3vl_30B.sh. - Updates pretrain_transformers.py to read ENABLE_GC_THRESHOLD. - When ENABLE_GC_THRESHOLD is set to "1", pretrain_transformers.py applies gc.set_threshold(700, 10, 1000). - If the environment variable is not set, the default behavior is unchanged. ## Does this PR introduce any user-facing change? Yes, but it is limited to an optional environment variable. Users can enable the GC threshold behavior by setting: export ENABLE_GC_THRESHOLD=1 The Qwen3-VL 30B LoRA example script enables this variable by default. Other training scripts are not affected unless they also set ENABLE_GC_THRESHOLD=1. No public API is changed. Related script: examples/qwen3vl/finetune_lora_qwen3vl_30B.sh ## 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!2593 | 3 个月前 |
| style: pre-commit autofix cleancode (base check) Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2616 merge master into master style: pre-commit autofix cleancode (base check) Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## 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!2616 | 3 个月前 |
| feat: upgrade diffusers Co-authored-by: htwang<wanghaitao60@huawei.com> # message auto-generated for no-merge-commit merge: !2766 merge master into master feat: upgrade diffusers Created-by: htwang Commit-by: htwang Merged-by: ascend-robot Description: ## What this PR does / why we need it? 升级diffusers版本到0.38.0 ## Does this PR introduce any user-facing change? 1、升级diffusers依赖版本 2、删掉日落模型部分代码 ## How was this patch tested? ST、UT看护验证 See merge request: Ascend/MindSpeed-MM!2766 | 2 个月前 |
