| Files | Last commit | Last update |
|---|---|---|
fix(conversion): resolve AICPU codecheck warnings Co-authored-by: hid29204727<wangboyu20@huawei.com> # message auto-generated for no-merge-commit merge: !5825 merge fix/conversion-aicpu-codecheck into master fix(conversion): resolve AICPU codecheck warnings Created-by: hid29204727 Commit-by: hid29204727 Merged-by: cann-robot Description: ## Description Resolve conversion AICPU codecheck findings that are safe to change without altering public behavior: - replace four reinterpret_cast conversions from void* with static_cast; - mark read-only TileWithAxis and TransData helpers as const; - align declaration parameter names with definitions; - make the read-only tensor parameter pointer-to-const; - remove an unused include. The two enum-class findings in formats_definitions.h are intentionally not changed because these dimension enums are used directly as integer indexes across the format-transfer module. The reported TransData reinterpret-cast locations are no longer present on current master and require no code change. ## Tests - Repository clang-format hook: passed. - trailing-whitespace hook: passed. - git diff --check: passed. - GitCode remote hooks: passed. - Full AICPU build/UT not run locally because CMake and the CANN build environment are unavailable on this Windows host. ## Type - [x] Bug fix - [ ] Feature - [ ] Documentation See merge request: cann/ops-math!5825 | 1 month ago | |
fix(aicpu): tile_with_axis 析构函数补充 override 修饰(G.CLS.12 残留) Co-authored-by: liu-wei<lovline.liuwei@huawei.com> # message auto-generated for no-merge-commit merge: !5920 merge fix/aicpu-cls12-destructor-override into master fix(aicpu): tile_with_axis 析构函数补充 override 修饰(G.CLS.12 残留) Created-by: liu-wei Commit-by: liu-wei Merged-by: cann-robot Description: ## 描述 修复 G.CLS.12 告警最后残留 1 处(重写虚函数应显式指定 override): | 告警 | 文件 | 修复 | |---|---|---| | G.CLS.12 | tile_with_axis_aicpu.h | ~TileWithAxisCpuKernel() override = default; | 基类 CpuKernel 析构为 virtual;同类其余 6 处(strided_slice/lin_space/top_k_pq_distance/top_k_v2_d/topk/select_v2)已在最新 master 修复,本提交补齐最后一处。 ## 验证 - override 为纯编译期检查,零代码生成影响;tile_with_axis_aicpu.cpp 语法级编译验证通过; - CI 流水线待跑。 ## 关联的Issue https://gitcode.com/cann/ops-math/issues/3279 ## 类型 🧹 代码清理 See merge request: cann/ops-math!5920 | 1 month ago | |
add aicpu:tile_with_axis&topkpqdistance Co-authored-by: sujunwei3<sujunwei3@huawei.com> # message auto-generated for no-merge-commit merge: !4174 merge dev into master add aicpu:tile_with_axis&topkpqdistance Created-by: sujunwei3 Commit-by: sujunwei3 Merged-by: cann-robot Description: ## 描述 新增四个 AI CPU 算子: 1. **tile_with_axis**(conversion):沿指定维度复制输入 Tensor 数据,扩展输出 Tensor。包含 AI Core 和 AI CPU 双实现。 2. **top_k_pq_distance**(math):基于分组极值剪枝的 TopK 距离计算,用于向量检索场景。支持动态多组输入。 3. **top_k_v2_d**(math):沿指定维度找出最大或最小的 k 个元素及其索引,兼容 TensorFlow TopKV2,包含 infershape 实现。 4. **zeta**(math):计算 Hurwitz zeta 函数 ζ(x, q) = Σ(n=0→∞) 1/(q+n)^x。 每个算子包含:op_kernel_aicpu 实现、op_graph 构图定义、examples 示例、UT 单元测试、README 文档。 ## 关联的Issue #2294 ## 测试 根据代码变更,测试场景如下: 1. **tile_with_axis 单元测试 + GE IR 验证** - 执行 bash build.sh -u --opkernel_aicpu --ops=tile_with_axis --soc=ascend910b,17 个 UT 全部通过 - 执行编包 + GE IR 运行,execute samples success,输出 bin 文件正常 2. **top_k_pq_distance 单元测试** - 执行 bash build.sh -u --opkernel_aicpu --ops=top_k_pq_distance --soc=ascend910b,5 个 UT 全部通过 3. **top_k_v2_d 单元测试 + GE IR 验证** - 执行 bash build.sh -u --opkernel_aicpu --ops=top_k_v2_d --soc=ascend910b,UT 全部通过 - 执行编包 + GE IR 运行,验证通过 4. **zeta 单元测试 + GE IR 验证** - 执行 bash build.sh -u --opkernel_aicpu --ops=zeta --soc=ascend910b,8 个 UT 全部通过 - 执行编包 + GE IR 运行,execute samples success,输出 bin 文件正常 ## 文档更新 - 新增 4 个算子 README.md(tile_with_axis、top_k_pq_distance、top_k_v2_d、zeta) - 更新 docs/zh/op_list.md,新增 4 个算子条目 ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [x] 文档更新 - [ ] 其他,请描述: See merge request: cann/ops-math!4174 | 3 months ago |
| Files | Last commit | Last update |
|---|---|---|
| 1 month ago | ||
| 1 month ago | ||
| 3 months ago |