Frost Mitchell

Intel Corporation

Results 13 comments of


                                            Frost Mitchell

rfc: propose split op for oneDNN Graph API

> @frost-intel @chunyuan-w, do you know how Dynamo handles such cases? Thanks! Dynamo can either use a constant input size similar to PT JIT, or it can have a dynamic...

[inductor][cpp] Add BMM kernel template for autotuning

A quick description of current changes and how comments have been addressed: ### BMM via calls to GEMM We use calls to the existing GEMM template to execute BMM efficiently...

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot merge

[inductor][cpp] Add BMM kernel template for autotuning

@jansel Could you review/approve this PR? A few files need approval from a core reviewer. Thanks.

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot merge

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot merge

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot rebase

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot merge

[inductor][cpp] Add BMM kernel template for autotuning

@pytorchbot merge

[Don't merge] [Kineto][submodule] Update kineto pin for XPU toggle feature

@pytorchbot merge

1
2
›