Stanley Winata 136021424c [SD] Change default VMA large heap block size for windows perf. (#715) ...

Windows perform can boost from 2.67s/image to 2.4523s/image.
While Linux stays the same.

2022-12-24 01:40:58 +07:00

..

examples

[SD] Change default VMA large heap block size for windows perf. (#715)

2022-12-24 01:40:58 +07:00

iree_utils

[Vulkan][utils] Add GTX Pascal support. (#709)

2022-12-22 15:24:15 -08:00

sharkdynamo

Add sharkdynamo which combines shark with torchdynamo.

2022-09-23 22:40:02 +05:30

tests

Add stress test

2022-12-13 13:21:51 -08:00

__init__.py

Modify the directory structure to remove conflicts.

2022-04-04 21:23:33 +05:30

backward_makefx.py

Add shark_importer for torch_models. (#183)

2022-07-12 20:38:19 -07:00

iree_eager_backend.py

Fix IREE eager backend device string (#237)

2022-08-03 12:09:52 -07:00

model_annotation.py

Modify model annotation tool to walk through ops by shape (#692)

2022-12-21 10:46:30 -08:00

parser.py

Move most xfails to entries in tank/all_models.csv and temporarily remove multiprocessing and TF gpu support. (#646)

2022-12-16 12:56:32 +05:30

shark_benchmark_runner.py

Move most xfails to entries in tank/all_models.csv and temporarily remove multiprocessing and TF gpu support. (#646)

2022-12-16 12:56:32 +05:30

shark_downloader.py

Add the download progress bar.

2022-12-21 15:47:33 +05:30

shark_importer.py

Update the models to 8th Dec version.

2022-12-14 00:01:46 +05:30

shark_inference.py

[SD] Add support for a compiled version of the discrete Euler scheduler (#657)

2022-12-17 19:25:43 -08:00

shark_runner.py

[SD] Add support for a compiled version of the discrete Euler scheduler (#657)

2022-12-17 19:25:43 -08:00

shark_trainer.py

Add shark_importer for torch_models. (#183)

2022-07-12 20:38:19 -07:00

stress_test.py

Fix concurrency issue in stress_test for CUDA devices

2022-12-22 08:54:19 -08:00

tflite_utils.py

Remove tf package dep for SharkDownloader tflite tests (#212)

2022-07-26 12:41:12 -07:00

torch_mlir_lockstep_tensor.py

[Lockstep] Hack to avoid aten._reshape_alias (#332)

2022-09-22 18:02:09 -04:00

torch_mlir_utils.py

Changes incorporating the recent torch_mlir compile api changes.

2022-11-15 15:25:37 +05:30