Basic operator performance optimization on PVC - Githubissues

intel / torch-xpu-ops

Apache License 2.0

30 stars 21 forks source link

Basic operator performance optimization on PVC #1108

Open xytintel opened 15 hours ago

xytintel commented 15 hours ago

🐛 Describe the bug

Evaluate the following operators in performance:

[x] batch_norm https://github.com/intel/torch-xpu-ops/pull/933
[ ] scatter_gather https://github.com/intel/torch-xpu-ops/pull/1112
[ ] grid_sampler
[ ] group_norm

Versions

Release/2.6