Ean Garvey
9035a2eed3
Add --local_tank_cache flag and update requirements. ( #368 )
...
* Add --local_tank_cache flag and update requirements.
* Update requirements-importer.txt
2022-09-28 03:02:59 -05:00
Ean Garvey
6cf5564c84
Remove "gpu" device alias and migrate to using "cuda" for NVIDIA GPU. ( #325 )
...
* Replace instances of "gpu" alias for devices with "cuda"
2022-09-13 01:16:56 -05:00
Ean Garvey
6aeda3670f
Split nightly workflow by backend (IREE / SHARK) ( #313 )
...
* Fix validation for nightly builds.
* Add option to generate shark_tank inside SHARK project
Add shark_arg for updating tank on mismatched hash (downloader)
* Fixup CI tank dir option.
* Fixup work directory variable
2022-09-09 22:51:30 +05:30
Daniel Garvey
d45a496030
adds a flag to enable directory choice ( #303 )
...
individual tests will require implementation of the flag
alternatively, simply passing shark_default_sha in your
individual app's download function will allow for this behavior
2022-08-31 22:17:40 -07:00
Ean Garvey
3514822cac
Improvements to pytest benchmarks. ( #267 )
...
* Add ONNX env var flags for venv setup.
* Setup arguments for ONNX benchmarking via pytest.
* Enable ONNX benchmarking on MiniLM via pytest (experimental)
* Fix sequence lengths to 128 for TF model creation and fix issue with benchmarks.
* Disable CI CPU benchmarks on A100, change some default args.
* add xfails for roberta TF model tests on GPU.
2022-08-17 02:29:48 -05:00
Ean Garvey
a8b021dc8d
Add benchmarks to MHLO miniLM and resnet50 and add dialect, num_iterations ( #264 )
2022-08-16 13:55:40 -05:00
Ean Garvey
23619068eb
Disable passing of sm_arch to iree-compile CL args by default. ( #253 )
...
* Disable passing of sm_arch to iree-compile CL args by default.
* Fix formatting.
2022-08-10 01:19:24 -07:00
Prashant Kumar
0dcf387089
Add shark_importer for torch_models. ( #183 )
...
All the torch_models are imported to gs::shark_tank.
Scripts have been updated.
2022-07-12 20:38:19 -07:00
Anush Elangovan
a7435973d9
Fix black formatting
2022-06-30 20:42:02 +00:00
Ean Garvey
0a6bc6e17f
Generate test-specific repro path for each TF model test. ( #158 )
...
Set TempFileSaver path directory to shark_args.repro_dir
2022-06-23 21:58:45 -07:00
Chi_Liu
fa0aaf63c2
Fix repo_dir cannot export/write mlir file bug ( #157 )
...
-Set repro_dir as a temporary dir within current working directory.
2022-06-23 18:37:16 -07:00
powderluv
a7ca9b8d68
Update parser.py
2022-06-23 15:26:17 -07:00
Prashant Kumar
e8aa105b2a
Divide iree_utils and do module imports on function calls.
2022-06-22 14:17:33 +05:30
Prashant Kumar
9d2ce33bd1
yapf format python files.
...
Please use `yapf -i --style .style.yapf shark/*.py` to format python
files.
2022-06-06 14:34:59 +05:30
Stanley Winata
c70ab55717
Add saving and loading of mlir to run. ( #89 )
2022-06-05 08:42:40 -07:00
yzhang93
16c50cac15
Add the option to use tuned model in shark_runner ( #79 )
2022-06-02 18:25:33 -07:00
Stanley Winata
5ece8f6f5d
Fix Test Issues on macos ( #57 )
...
-Added a separate pytest config for macos
-Adjusted shark parser to ignore unknown args
2022-05-24 14:40:35 -07:00
Prashant Kumar
a280b0a651
Add decompositions for bert_training.
2022-05-09 16:39:11 +05:30
Prashant Kumar
08df345fbc
Add make_fx whole graph extract.
2022-05-04 20:58:59 +05:30
Vivek Khandelwal
7f6159462d
Add num_iterations and num_warmup_iterations args.
2022-05-04 20:38:19 +05:30
Prashant Kumar
db5be15310
Centralize the parser location. Also add the --device flag.
...
Centralized the shark_args parser. Also added the --device flag that
specifies the device on which the inference or training is to be done.
2022-05-02 15:35:22 +05:30
