master

ascend-robot【feature】 支持FA3的混合量化配置及保存

dfbca739创建于 3 小时前历史提交

文件	最后提交记录	最后更新时间
app	【UT】补充core目录UT Co-authored-by: joejoezhou<zhourongchen1@huawei.com> # message auto-generated for no-merge-commit merge: !487 merge feature/core-ut into master 【UT】补充core目录UT Created-by: joejoezhou Commit-by: joejoezhou Merged-by: ascend-robot Description: ## PR描述 (What this PR does / why we need it?) 补充msmodelslim/core单测 ## 面向用户的变更 (Does this PR introduce _any_ user-facing change)? 无 ## 功能验证 (How was this patch tested?) - [x] 功能自验 - [x] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [x] 新增/变更内容是否已新增/适配UT测试用例看护 See merge request: Ascend/msmodelslim!487	20 小时前
cli	【test】 补全 cli/app/flat_quant 三模块单测，达 80% 行覆盖门禁 Co-authored-by: rookie_hongchuan<hongchuan6@h-partners.com> # message auto-generated for no-merge-commit merge: !471 merge feature/ut-coverage into master 【test】 补全 cli/app/flat_quant 三模块单测，达 80% 行覆盖门禁 Created-by: rookie_hongchuan Commit-by: rookie_hongchuan Merged-by: ascend-robot Description: ## PR描述 (What this PR does / why we need it?) 补全 msmodelslim/cli + app + processor/flat_quant 三模块单测覆盖，达到行覆盖率 80% / 分支覆盖率 60% 门禁。 新增 21 个单测文件，覆盖正常/边界/异常三类情形，模块聚合行覆盖 85% / 分支覆盖 80%（全部达标）。 ## 面向用户的变更 (Does this PR introduce _any_ user-facing change?) 无。仅新增单测，不修改业务源码。 ## 功能验证 (How was this patch tested?) - [x] 功能自验：pytest test/cases/app/ test/cases/cli/ test/cases/processor/flat_quant/ test/smoke/ → 402 passed, 48 skipped, 3 xfailed - [x] 覆盖率：三模块行覆盖 81%-98%，均过 80% 门禁 ### 覆盖明细 | 模块 | 行覆盖 | 分支覆盖 | |------|--------|---------| | msmodelslim/cli/ | 98% | 95% | | msmodelslim/processor/flat_quant/ | 84% | 75% | | msmodelslim/app/ | 81% | 76% | ### 新增单测文件（21 个） - test/cases/cli/ — test_utils.py, test_cli_main.py, test_naive_main.py, test_analysis_main.py, test_tuning_main.py - test/cases/app/auto_tuning/ — test_application.py, test_evaluation_service_infra.py, test_model_info_interface.py, test_plan_manager_infra.py, test_practice_accuracy_infra.py, test_practice_history_infra.py, test_practice_manager_infra.py - test/cases/app/naive_quantization/ — test_application_helpers.py - test/cases/processor/flat_quant/flat_quant_utils/ — test_fqu_utils.py, test_trans_matrix.py, test_flat_quant_manager.py, test_fake_clip_quantizer.py, test_flat_fake_quant_linear.py, test_structure_pair.py, test_structure_pair_real.py - test/smoke/ — test_flat_quant.py See merge request: Ascend/msmodelslim!471	23 小时前
common	【msmodelslim】刷新文件头部声明和版权 Co-authored-by: caishengcheng<caishengcheng@huawei.com> # message auto-generated for no-merge-commit merge: !28 merge license into master 【msmodelslim】刷新文件头部声明和版权 Created-by: caishengcheng Commit-by: caishengcheng Merged-by: ascend-robot Description: 【msmodelslim】刷新文件头部声明和版权 See merge request: Ascend/msmodelslim!28	4 个月前
core	【feature】 支持FA3的混合量化配置及保存 Co-authored-by: wangsihao<wangsihao5@h-partners.com> # message auto-generated for no-merge-commit merge: !483 merge master into master 【feature】 支持FA3的混合量化配置及保存 Created-by: wangsihao Commit-by: wangsihao Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 ## PR描述 (What this PR does / why we need it?) 支持FA3量化中的不同策略和粒度的混合配置 如：fa_q 动态量化，fa_k/fa_v 静态量化 量化粒度支持FP8/INT8，其中INT8仅支持静态量化 ## 功能验证 (How was this patch tested?) - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 See merge request: Ascend/msmodelslim!483	3 小时前
format	【msmodelslim】[feature] support compressed_tensors format Co-authored-by: anreywmh<18845895998@163.com> # message auto-generated for no-merge-commit merge: !451 merge quantFormatRefactor/compressed_tensors into master 【msmodelslim】[feature] support compressed_tensors format Created-by: anreywmh Commit-by: anreywmh Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 Thanks for sending a pull request! BEFORE SUBMITTING, PLEASE READ [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md). ## PR描述 (What this PR does / why we need it?) - 请明确说明您提交PR的变更内容。本部分旨在概述所做的变更，以及此PR是如何解决该问题的。请尽可能地提供有助于评审人员更高效、更快速完成检视审查的实用说明。 在 V1 框架中首次完整打通 compressed-tensors 落盘链路：逐层写 safetensors 权重 + 收尾写 config.json 量化元数据 + 配套 schema/反向推导模块 + 文档，使导出结果可直接用于 vLLM compressed-tensors 推理 1、实现 compressed-tensors 完整导出：将 CompressedTensorsQuantFormat 从空壳补全为可落盘实现，支持 W8A8 静态/动态量化层写 int8 权重及 scale 等参数，未量化层写 float 权重，输出与 vLLM compressed-tensors 对齐的 safetensors。 2、收尾写入 config.json：在 finalize_export 中复制源模型 HF 附属文件，并从 QIR 模型反向推导 quantization_config 写入 config.json，供推理框架识别量化方案。 3、新增 quantization 子模块：补充 QuantizationScheme、QuantizationConfig、quant_config_builder 等，定义 QIR 层与 compressed-tensors preset 的映射，以及 config_groups、ignore 等元数据的自动推断。 4、重构 QuantFormatBase 遍历逻辑：统一 WrapperIR 处理与 handler 分发（精确类型匹配），为多种落盘格式共用模块遍历能力。 5、适配 QuantSaveProcessor：ExportContext 路径改为 Path 类型，与格式层接口一致。 6、补充文档：新增 compressed_tensors.md 端到端说明，并在一键量化 usage 中增加该格式的配置入口。 - 请说明为何需要这些更改，例如具体的使用场景或bug描述。 - 关联issue号（如果有）。 - Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue. If possible, please consider writing useful notes for better and faster reviews in your PR. - Please clarify why the changes are needed. For instance, the use case and bug description. - Related issue number (if any) ## 面向用户的变更 (Does this PR introduce _any_ user-facing change)? - 请注意，这里指的是**任何**面向用户的变更，包括但不限于API、用户界面或其他使用方式上的变更。 - Note that it means *any* user-facing change including all aspects such as API, interface or other behavior changes. ## 功能验证 (How was this patch tested?) 请确认CI已通过增量及存量的单元测试用例。 如果本次测试方式与常规单元测试不同，请详细说明您的测试步骤(最好提供完整的可复现的操作路径及关键截图)，以便Committer能够快速复现验证，也便于后续的维护。 如果未添加测试，请说明未添加的原因，以及为何难添加测试。 - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 CI passed with new added/existing test. If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future. If tests were not added, please describe why they were not added and/or why it was difficult to add. - [_] Self-verification of the feature. - [_] Screenshot of local self-verification (please anonymize any sensitive information such as personal identifiers) - [_] Have new or modified unit test (UT) cases been added or adapted to cover the newly added or changed content? See merge request: Ascend/msmodelslim!451	2 天前
infra	【bugfix】自动调优计算yaml配置md5问题修复 Co-authored-by: xyxin_006<xyxin_hit@163.com> # message auto-generated for no-merge-commit merge: !420 merge bugfix/auto_tune into master 【bugfix】自动调优计算yaml配置md5问题修复 Created-by: xyxin_006 Commit-by: xyxin_006 Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 Thanks for sending a pull request! BEFORE SUBMITTING, PLEASE READ [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md). ## PR描述 (What this PR does / why we need it?) 问题：自动调优精度复用机制是基于yaml配置的md5值，当前的md5值计算会考虑yaml的 metadata，其中config_id会随着迭代次数而变化，从而导致一样的量化配置计算出的md5值不一致，无法复用历史精度数据。 修复后：自动调优中计算practice的md5值不考虑 metadata ## 面向用户的变更 (Does this PR introduce _any_ user-facing change)? - 请注意，这里指的是**任何**面向用户的变更，包括但不限于API、用户界面或其他使用方式上的变更。 - Note that it means *any* user-facing change including all aspects such as API, interface or other behavior changes. ## 功能验证 (How was this patch tested?) 请确认CI已通过增量及存量的单元测试用例。 如果本次测试方式与常规单元测试不同，请详细说明您的测试步骤(最好提供完整的可复现的操作路径及关键截图)，以便Committer能够快速复现验证，也便于后续的维护。 如果未添加测试，请说明未添加的原因，以及为何难添加测试。 - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 CI passed with new added/existing test. If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future. If tests were not added, please describe why they were not added and/or why it was difficult to add. - [_] Self-verification of the feature. - [_] Screenshot of local self-verification (please anonymize any sensitive information such as personal identifiers) - [_] Have new or modified unit test (UT) cases been added or adapted to cover the newly added or changed content? See merge request: Ascend/msmodelslim!420	13 天前
ir	【feature】 支持FA3的混合量化配置及保存 Co-authored-by: wangsihao<wangsihao5@h-partners.com> # message auto-generated for no-merge-commit merge: !483 merge master into master 【feature】 支持FA3的混合量化配置及保存 Created-by: wangsihao Commit-by: wangsihao Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 ## PR描述 (What this PR does / why we need it?) 支持FA3量化中的不同策略和粒度的混合配置 如：fa_q 动态量化，fa_k/fa_v 静态量化 量化粒度支持FP8/INT8，其中INT8仅支持静态量化 ## 功能验证 (How was this patch tested?) - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 See merge request: Ascend/msmodelslim!483	3 小时前
mindspore	【msmodelslim】刷新文件头部声明和版权 Co-authored-by: caishengcheng<caishengcheng@huawei.com> # message auto-generated for no-merge-commit merge: !28 merge license into master 【msmodelslim】刷新文件头部声明和版权 Created-by: caishengcheng Commit-by: caishengcheng Merged-by: ascend-robot Description: 【msmodelslim】刷新文件头部声明和版权 See merge request: Ascend/msmodelslim!28	4 个月前
model	【feature】 支持FA3的混合量化配置及保存 Co-authored-by: wangsihao<wangsihao5@h-partners.com> # message auto-generated for no-merge-commit merge: !483 merge master into master 【feature】 支持FA3的混合量化配置及保存 Created-by: wangsihao Commit-by: wangsihao Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 ## PR描述 (What this PR does / why we need it?) 支持FA3量化中的不同策略和粒度的混合配置 如：fa_q 动态量化，fa_k/fa_v 静态量化 量化粒度支持FP8/INT8，其中INT8仅支持静态量化 ## 功能验证 (How was this patch tested?) - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 See merge request: Ascend/msmodelslim!483	3 小时前
onnx	【msmodelslim】刷新文件头部声明和版权 Co-authored-by: caishengcheng<caishengcheng@huawei.com> # message auto-generated for no-merge-commit merge: !28 merge license into master 【msmodelslim】刷新文件头部声明和版权 Created-by: caishengcheng Commit-by: caishengcheng Merged-by: ascend-robot Description: 【msmodelslim】刷新文件头部声明和版权 See merge request: Ascend/msmodelslim!28	4 个月前
processor	【feature】 支持FA3的混合量化配置及保存 Co-authored-by: wangsihao<wangsihao5@h-partners.com> # message auto-generated for no-merge-commit merge: !483 merge master into master 【feature】 支持FA3的混合量化配置及保存 Created-by: wangsihao Commit-by: wangsihao Merged-by: ascend-robot Description: 感谢您贡献的Pull Request！ 在提交之前，请务必阅读 [CONTRIBUTING.md](https://gitcode.com/Ascend/msmodelslim/blob/master/CONTRIBUTING.md)。 ## PR描述 (What this PR does / why we need it?) 支持FA3量化中的不同策略和粒度的混合配置 如：fa_q 动态量化，fa_k/fa_v 静态量化 量化粒度支持FP8/INT8，其中INT8仅支持静态量化 ## 功能验证 (How was this patch tested?) - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 See merge request: Ascend/msmodelslim!483	3 小时前
pytorch	[security] fix some security scan issue Co-authored-by: zhongzhoutan<1710115119@bjmu.edu.cn> # message auto-generated for no-merge-commit merge: !235 merge security/release_acl_explicit into master [security] fix some security scan issue Created-by: tangxuanya Commit-by: zhongzhoutan Merged-by: ascend-robot Description: [security] use command list instead of str format to avoid invalid split when parameters have space 1. string format will cause parameter invalidation when parameter has space for subprocess run See merge request: Ascend/msmodelslim!235	2 个月前
quant	【msmodelslim】刷新文件头部声明和版权 Co-authored-by: caishengcheng<caishengcheng@huawei.com> # message auto-generated for no-merge-commit merge: !28 merge license into master 【msmodelslim】刷新文件头部声明和版权 Created-by: caishengcheng Commit-by: caishengcheng Merged-by: ascend-robot Description: 【msmodelslim】刷新文件头部声明和版权 See merge request: Ascend/msmodelslim!28	4 个月前
utils	【feature】预检失败由error改为warning，后续适配器报错再提供版本依赖 Co-authored-by: libarry<870390541@qq.com> # message auto-generated for no-merge-commit merge: !485 merge bugfix/precheck into master 【feature】预检失败由error改为warning，后续适配器报错再提供版本依赖 Created-by: libarry Commit-by: libarry Merged-by: ascend-robot Description: ## PR描述 (What this PR does / why we need it?) 1、依赖检查报错改为警告，所有版本依赖改为==或范围约束 2、当依赖检查失败后，适配器方法增加装饰器exception_handler，确保后续使用过程的适配器内任何报错会提示安装推荐的依赖版本号。 3、由于TimeoutError和内置TimeoutError重名，改名为ModelslimTimeoutError 4、修改UnexpectedError，增加两个方法，分别是注入tips和清理tips，支持提前注入外部Tips的能力，应用于warning后报错场景 ## 面向用户的变更 (Does this PR introduce _any_ user-facing change)? 用户环境依赖不符时，直接报错改为警告 ## 功能验证 (How was this patch tested?) 模型适配器问题注入tips:  模型forward报错触发UnexpectedError注入tips:  请确认CI已通过增量及存量的单元测试用例。 如果本次测试方式与常规单元测试不同，请详细说明您的测试步骤(最好提供完整的可复现的操作路径及关键截图)，以便Committer能够快速复现验证，也便于后续的维护。 如果未添加测试，请说明未添加的原因，以及为何难添加测试。 - [_] 功能自验 - [_] 本地自验截图(涉及个人标识符等敏感信息请注意脱敏) - [_] 新增/变更内容是否已新增/适配UT测试用例看护 CI passed with new added/existing test. If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future. If tests were not added, please describe why they were not added and/or why it was difficult to add. - [_] Self-verification of the feature. - [_] Screenshot of local self-verification (please anonymize any sensitive information such as personal identifiers) - [_] Have new or modified unit test (UT) cases been added or adapted to cover the newly added or changed content? See merge request: Ascend/msmodelslim!485	3 小时前
__init__.py	【msmodelslim】刷新文件头部声明和版权 Co-authored-by: caishengcheng<caishengcheng@huawei.com> # message auto-generated for no-merge-commit merge: !28 merge license into master 【msmodelslim】刷新文件头部声明和版权 Created-by: caishengcheng Commit-by: caishengcheng Merged-by: ascend-robot Description: 【msmodelslim】刷新文件头部声明和版权 See merge request: Ascend/msmodelslim!28	4 个月前