Enable BatchMatMul->FC optimization for i8. (989655b9) · Commits · gitlab-org / build / omnibus-mirror / tensorflow · JiHu GitLab

Skip to content

GitLab

Explore

Sign in

Primary navigation

Project

T

tensorflow
- Activity
- Members
- Labels
- Issues
  0
- Issue boards
- Milestones
- Iterations
- Wiki
- Requirements
- Environments
- Terraform modules
- Incidents
- Service Desk

Snippets Groups Projects

gitlab-org
build
omnibus-mirror
tensorflow
Commits
989655b9

Commit 989655b9 authored 4 months ago by

Weiyi Wang Committed by TensorFlower Gardener 4 months ago

Downloads
- Patches
- Plain Diff

Enable BatchMatMul->FC optimization for i8.

PiperOrigin-RevId: 731410805

parent bd1811ec

Changes 4

Hide whitespace changes

Inline Side-by-side

Showing

tensorflow/compiler/mlir/lite/tests/optimize_batch_matmul.mlir 1 addition, 1 deletion

...rflow/compiler/mlir/lite/tests/optimize_batch_matmul.mlir
tensorflow/compiler/mlir/lite/transforms/optimize_batch_matmul_pass.cc 6 additions, 7 deletions

...mpiler/mlir/lite/transforms/optimize_batch_matmul_pass.cc
third_party/xla/xla/service/copy_insertion_test.cc 79 additions, 0 deletions

third_party/xla/xla/service/copy_insertion_test.cc
third_party/xla/xla/tests/collective_pipeline_parallelism_test.cc 30 additions, 0 deletions

...rty/xla/xla/tests/collective_pipeline_parallelism_test.cc

with 116 additions and 8 deletions

0% or .

You are about to add 0 people to the discussion. Proceed with caution.

Finish editing this message first!

Please register or sign in to comment