master

ascend-robotadd npu_format_cast_via_cpu + add quantmatmul lite converter

04e03b69创建于 6 天前历史提交

文件	最后提交记录	最后更新时间
pyproject.toml	修改完善pre-commit开源代码检测工具 Co-authored-by: wujinyuan1<wujinyuan1@huawei.com> # message auto-generated for no-merge-commit merge: !3023 merge master into master 修改完善pre-commit开源代码检测工具 Created-by: wujinyuan1 Commit-by: wujinyuan1 Merged-by: ascend-robot Description: 修改完善pre-commit开源代码检测工具 See merge request: Ascend/torchair!3023	24 天前
typos.toml	add npu_format_cast_via_cpu + add quantmatmul lite converter # message auto-generated for no-merge-commit merge: !3092 merge master into master add npu_format_cast_via_cpu + add quantmatmul lite converter Created-by: zihan0007 Commit-by: zihan0007 Merged-by: ascend-robot Description: 合入来源： torchair 新增npu_format_cast_via_cpu接口： 新增npu_format_cast_via_cpu接口 支持npuTensor在cpu将数据排布从Nd转为Nz，当前支持int8/uint8 Nd输入转为Fractal_nz输出 ge_pass增加QuantBatchMatmulV3前的Bitcast和Transpose算子互换位置的能力，方便后续Transpose融合 增加lite converter支持QuantBatchMatmulV3 A8W4（int4）场景的converter逻辑 See merge request: Ascend/torchair!3092	6 天前