yankf21
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
11,477 Commits 18 Branches 22 Tags 3.6 GiB
Commit Graph

9 Commits

Author SHA1 Message Date
tronzhang c32bf5ac28 promote complex tensor as graph's input and recorrect getitem index for graph kernels fusion. 2020-10-10 17:57:42 +08:00
dayschan 7599686a72 GraphKernel supports multi-output kernels 2020-09-30 09:27:44 +08:00
lingyunli63 dd48f10c3d add assign ops in composite_topi 2020-09-28 10:20:53 +08:00
dayschan 3c2da3197f Fix review_bot and codedex problems 2020-09-15 11:32:40 +08:00
mindspore-ci-bot f7691335eb !6167 fused select and greater op to improve bert perfermance on GPU 
Merge pull request !6167 from ZengZitao/fuse_greater_select_ms
 2020-09-14 20:25:20 +08:00
zengzitao a38d6139fa fused select and greater op to improve bert perfermance on GPU 2020-09-14 15:03:00 +08:00
r1chardf1d0 88de0cffa9 open graph kernel expander opt for gpu 2020-09-14 11:37:13 +08:00
mindspore-ci-bot 7152fe04be !5783 GraphKernel supports GPU 
Merge pull request !5783 from DeshiChen/graph_kernel_1.0
 2020-09-10 09:15:21 +08:00
dayschan 37a48f6aac GraphKernel supports GPU 
1. Update akg submodule
2. Refactor akg_kernel_build, akg_ascend_kernel_build, akg_gpu_kernel_build
3. Add akg_kernel_json_decoder to support converting kernel_json to AnfNode.
4. Add GraphKernel Cost Model. (mindspore/_extends/graph_kernel)
5. Add some GraphKernel passes to GpuSession, move these passes to backend/optimizer/graph_kernel.
6. Add global id for ir files.
7. Fix bug in ConstInputToAttr.
 2020-09-09 17:46:50 +08:00