Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Revert PR #48063: Add _FusedBatchNormGrad to support side_input and a…
…ctivation Still hits "unsupported cudnn" error To repro, run (from tensorflow_models repository): ``` //tensorflow_models/official/benchmark:keras_imagenet_benchmark -- --logtostderr --benchmarks=Resnet50KerasBenchmarkSynth.benchmark_xla_1_gpu_fp16\$ ``` PiperOrigin-RevId: 372380247 Change-Id: Iae61fa8095b0c60716cfdf147544d8f7b3e3f493
- Loading branch information
1 parent
edf29e8
commit 3eeb732
Showing
13 changed files
with
138 additions
and
905 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
3eeb732
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@cheshire Could you please share more info on how to repro the "unsupported cudnn" error. Like the command lines and GPUs.
I have tried to run the tf models benchmark on my V100 but cannot repro the issue. Below is how I did it and maybe I missed something:
/home/workspace/models
contains the tensorflow models repo with "r2.4" branch.What is inside test.py:
3eeb732
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Or if possible can you help check if the issue persists with this #48893.