| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
add support for step_audio2_mini model based on vLLM and vLLM-Ascend Co-authored-by: XiyangLiu233<liuxiyang10@huawei.com> Co-authored-by: LiuZhaoyuan<liuzhaoyuan6@huawei.com> # message auto-generated for no-merge-commit merge: !7603 merge master into master add support for step_audio2_mini model based on vLLM and vLLM-Ascend Created-by: frankliu233 Commit-by: frankliu233;LiuZhaoyuan;XiyangLiu233 Merged-by: ascend-robot Description: ## Motivation add support for step_audio2_mini model based on vLLM and vLLM-Ascend ## Modification add support code for step_audio2_mini model ## 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/ModelZoo-PyTorch!7603 | 12 天前 | |
arcface模型推理适配 Co-authored-by: lipan0927<2125524263@qq.com> # message auto-generated for no-merge-commit merge: !7635 merge arcface into master arcface模型推理适配 Created-by: lipan0927 Commit-by: lipan0927 Merged-by: ascend-robot Description: ## Motivation 适配arcface模型 ## Modification 提供arcface模型适配的源码和readme文件。 ## PR信息 [[Feature]: arcface模型推理适配](https://gitcode.com/Ascend/ModelZoo-PyTorch/issues/2693) ## Self-test (Optional) 自测通过，见自验报告 ## 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/ModelZoo-PyTorch!7635 | 1 小时前 | |
【fix】完善小模型Atlas 300I DUO硬件描述 Co-authored-by: Niushiya<niushiya1@huawei.com> # message auto-generated for no-merge-commit merge: !7587 merge master into master 【fix】完善小模型Atlas 300I DUO硬件描述 Created-by: niushiya Commit-by: Niushiya Merged-by: ascend-robot Description: ## Motivation 1、完善小模型Atlas 300I DUO硬件描述，补充单芯字段； ## 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/ModelZoo-PyTorch!7587 | 1 个月前 | |
[feature]Isaac-GR00T模型支持200I PRO设备推理；优化推理性能 Co-authored-by: frost_mourne<suyuxuan3@h-partners.com> # message auto-generated for no-merge-commit merge: !7634 merge master into master [feature]Isaac-GR00T模型支持200I PRO设备推理；优化推理性能 Created-by: frost_mourne Commit-by: frost_mourne Merged-by: ascend-robot Description: ## Motivation 支持200I PRO设备推理； 优化推理性能 ## Modification 增加了200I PRO设备不亲和操作的优化及替换。 部分静态数据缓存，如randn 部分算子替换，如conv3d、pfa算子 增加了多条轨迹分进程推理并存储结果的逻辑 结果输出逻辑提为函数，多进程推理结束后用已存储的结果输出最终结果。 ## 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**: - [ ] 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/ModelZoo-PyTorch!7634 | 19 天前 | |
tabPFN npu Co-authored-by: yuanxinzhuo15<yuanxinzhuo@huawei.com> # message auto-generated for no-merge-commit merge: !7625 merge tabPFN into master tabPFN npu Created-by: yuanxinzhuo15 Commit-by: yuanxinzhuo15 Merged-by: ascend-robot Description: ## Motivation ModelZoo 缺少 TabPFN 在昇腾 NPU 上的推理交付。本 PR 新增 TabPFN NPU 分类推理指导与自验脚本。 ## Modification 新增目录 ACL_PyTorch/built-in/foundation_models/TabPFN_for_Pytorch/： - README.md：环境、镜像、安装、补丁、自验、FAQ - requirements.txt：依赖（tabpfn==2.0.9 等） - apply_tabpfn_npu_patch.py：补丁 pip 包内 multi_head_attention.py，绕过 CUDA capability 检测 - tabpfn_npu_infer.py：sklearn 三数据集分类自验，对比 GPU 参考精度 ## Self-test | 项目 | 配置 | | :----------------------- | :----------------------------- | | 硬件 | Atlas 800I A2 单卡 | | CANN / torch / torch_npu | 8.3.RC1 / 2.1.0 / 2.1.0.post17 | | tabpfn | 2.0.9 | 结果（热启动，权重已缓存）： | 数据集 | 总耗时(s) | NPU accuracy | GPU 参考 | | :------------ | :-------- | :----------- | :------- | | breast_cancer | 2.71 | 0.9720 | 0.974 | | iris | 2.53 | 0.9737 | 1.000 | | wine | 2.65 | 1.0000 | 0.972 | 三集精度偏差均 ≤ 0.03；NPU 耗时约为 GPU 参考 3～9 倍（transfer_to_npu 路径预期表现）。 ## BC-breaking ## 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/ModelZoo-PyTorch!7625 | 18 天前 | |
chemberta模型推理适配 Co-authored-by: lipan0927<2125524263@qq.com> # message auto-generated for no-merge-commit merge: !7637 merge chemberta into master chemberta模型推理适配 Created-by: lipan0927 Commit-by: lipan0927 Merged-by: ascend-robot Description: ## Motivation 适配chemberta模型 ## Modification 提供chemberta模型适配的源码和readme文件。 ## PR信息 [[Feature]: chemberta模型适配 ](https://gitcode.com/Ascend/ModelZoo-PyTorch/issues/2691) ## Self-test (Optional) 自测通过，见自验报告 ## 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/ModelZoo-PyTorch!7637 | 1 天前 | |
fix PaddleOCR-VL-1.5, PP-DocLayoutV2, PP-DocLayoutV3 Co-authored-by: wangzihan-zzz<1780093255@qq.com> # message auto-generated for no-merge-commit merge: !7594 merge master into master fix PaddleOCR-VL-1.5, PP-DocLayoutV2, PP-DocLayoutV3 Created-by: wangzihan-zzz Commit-by: wangzihan-zzz Merged-by: ascend-robot Description: ## Motivation fix PaddleOCR-VL-1.5, PP-DocLayoutV2, PP-DocLayoutV3 docs ## Modification fix PaddleOCR-VL-1.5 README contents，update PP-DocLayoutV2 README、infer.py，fix PP-DocLayoutV3 README ## 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/ModelZoo-PyTorch!7594 | 1 个月前 | |
init | 4 年前 | |
init | 4 年前 | |
init | 4 年前 | |
init | 4 年前 |
华为昇腾 Modelzoo 推理模型仓库
支持Pytorch及ONNX模型