github/SHARK-Studio
1
1
Fork 0
mirror of https://github.com/nod-ai/SHARK-Studio.git synced 2026-01-09 05:47:54 -05:00
Code Issues Packages Projects Releases Wiki Activity
1,750 Commits 66 Branches 705 Tags
main
Commit Graph

222 Commits

Author SHA1 Message Date
pdhirajkumarprasad
f6dd02fa67 Fix for migration (#2183) 
Signed-off-by: pdhirajkumarprasad <dhirajp@amd.com>
 2025-11-24 15:32:02 +05:30
pdhirajkumarprasad
fe03539901 Migration to AMDShark (#2182) 
Signed-off-by: pdhirajkumarprasad <dhirajp@amd.com>
 2025-11-20 12:52:07 +05:30
Marius Brehler
6b7903793a Update links after repository renamings (#2174) 2024-10-17 18:37:30 +02:00
Ean Garvey
ed3dda94c0 Cleanup xfails in pytest suite. (#1995) 2023-11-29 23:16:15 -06:00
Sungsoon Cho
51e1bd1c5d (OPT) Fix typo in the message; s/reponse/response (#1920) 2023-11-09 11:00:48 -06:00
Ean Garvey
98244232dd Add smoothquant OPT to examples. (#1922) 2023-10-27 12:32:12 -05:00
PhaneeshB
679a452139 fix calls and remove unused imports for check_device_drivers 2023-10-27 10:30:40 +05:30
Sungsoon Cho
0c38c33d0a Add opt_causallm_samples.py. (#1916) 2023-10-25 11:52:51 -05:00
Ean Garvey
841773fa32 Updates to opt_causallm example (#1905) 
* Updates to opt_causallm example

* Fixup opt_perf_comparison.py

* Use same filenames across opt examples.
 2023-10-24 10:54:39 -07:00
Ean Garvey
e6cb5cef57 Add --additional_runtime_args option and use in OPT example. (#1855) 
* Add --additional_runtime_args option and use in OPT example.

Fix the func name. (#1838)

Co-authored-by: Sungsoon Cho <sungsoon.cho@gmail.com>
 2023-10-19 13:29:39 -05:00
Ean Garvey
caf6cc5d8f Switch most compile flows to use ireec.compile_file. (#1863) 
* Switch most compile flows to use ireec.compile_file.

* re-add input type to compile_str path.

* Check if mlir_module exists before checking if it's a path or pyobject.

* Fix some save_dir cases
 2023-10-06 23:04:43 -05:00
Ean Garvey
8614a18474 Remove tf dependencies from importer path. (#1874) 
* Remove tf dependencies from import path.

* Fix formatting.
 2023-10-06 12:27:12 -07:00
Sungsoon Cho
0d16c81687 Remove unused import. (#1857) 2023-10-02 11:36:08 -05:00
Sungsoon Cho
b5fbb1a8a0 Rename the func arg save_json to avoid name collision. (#1837) 
* Rename the func arg save_json to avoid name collision.

* black formatted.
 2023-09-19 17:29:27 -05:00
Abhishek Varma
87ae14b6ff [SD] Add sdpfa decomposition + update IREE flag 
-- This commit adds Scaled Dot Product Flash Attention's decomposition
   in shark_importer.
-- It also updates `iree-flow-enable-data-tiling` to `iree-opt-data-tiling`.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-09-04 18:03:53 +05:30
Ean Garvey
9697981004 Pipe through a debug option to iree compile utils. (#1796) 
* Update compile_utils.py

* Pipe through a flag to toggle debug options in compile utils.

* Update SharkLLMBase.py
 2023-08-25 07:11:11 -07:00
Ean Garvey
07f6f4a2f7 Add a short README for the OPT examples and small tweaks. (#1793) 
* Small changes to OPT example.

* Update opt README.

* Add a few modes to batch script.

* Update README.md
 2023-08-24 17:26:11 -07:00
Ean Garvey
8e3860c9e6 Remove flags that are default in upstream IREE (#1785) 
* Remove index bits flags now set by default

* Update shark_studio_imports.py
 2023-08-24 11:57:54 -05:00
Sungsoon Cho
79075a1a07 Opt perf (#1786) 
* Define command line args, model-name, max-seq-len, platform, etc.

* Add usage example.

* Add opt_perf_comparision_batch.py.

* Use shlex instead.
 2023-08-24 08:33:12 -05:00
Ean Garvey
cb509343d9 Fix pytest benchmarks and shark_tank generation. (#1632) 
- fix setup_venv.sh for benchmarks/imports etc.
- fix torch benchmarks in SharkBenchmarkRunner
- generate SD artifacts using build_tools/stable_diffusion_testing.py and --import_mlir
- decouple SD gen from tank/generate_sharktank for now
 2023-08-16 17:48:47 -05:00
Ean Garvey
c96571855a prevents recompiles for cuda benchmarks + update benchmark_module path (#1759) 
* xfail resnet50_fp16

* Fix cuda benchmarks and prevent recompilation.
 2023-08-14 15:30:32 -05:00
Ean Garvey
a9f95a218b Remove SD from all_models.csv (#1706) 
Removes SD from pytests as it has its own test suite.
 2023-08-05 15:55:52 -05:00
Ean Garvey
dd738a0e02 small changes to opt_perf_comparison.py (#1670) 
* Use longer prompts for OPT comparison script

* small tweaks
 2023-07-19 06:26:50 -07:00
Ean Garvey
7cf19230e2 add perf comparison script for opt. (#1650) 2023-07-13 13:29:48 -05:00
Ean Garvey
7dcbbde523 Xfail models for data tiling flag changes (#1624) 2023-07-06 06:57:17 -07:00
Ean Garvey
6e6c90f62b Pin torch-mlir and use local-task in OPT. (#1592) 2023-06-23 19:17:05 -07:00
Chi_Liu
f60ab1f4fa Add Deberta to stablehlo in shark tank (#1545) 2023-06-16 13:24:44 -07:00
Ean Garvey
eb6d11cfed Change mlir dialects for tf tests to stablehlo. (#1535) 
* Change mlir dialects for tf tests to stablehlo

* Update shark_runner.py
 2023-06-14 10:43:49 -07:00
Ean Garvey
49b3ecdbca (pytest) don't run redundant tests in cpu suite (#1532) 2023-06-13 22:40:33 -05:00
Ean Garvey
f53e3594c3 OPT Refactor (#1516) 
* Change script to 1.3b model and add pytorch comparison

* fix CLI command

* Match OPT transformers model updates + numerics against latest version

* Cleanup OPT sentence completion script.

* Fix formatting and add standalone validation scripts.

* Add minimal OPT wrapper and example with import_with_fx

* Rename OPT full model wrapper.

* Cleanup test scripts for OPT.
 2023-06-13 22:40:07 -05:00
Ean Garvey
5562d1dfda Fix xfails for cpu pytest cases (#1527) 
Adding cpu-sync and cpu-task device configs was allowing respective tests to bypass the xfail conditional for cpu pytests marked in tank/all_models.csv. This commit updates the conditional to xfail those cases for cpu-sync and cpu-task as well.
 2023-06-13 17:01:51 -07:00
Ean Garvey
2c3485ca3e Add standalone OPT sentence completion script. (#1506) 2023-06-07 10:58:03 -07:00
Ean Garvey
a07fe80530 Update OPT, ResNet example scripts. (#1492) 
* Update API in OPT example.

* fix resnet50 script

* Add OPT1.3b test script.
 2023-06-05 20:19:35 -07:00
Ean Garvey
4151ec3a8f (pytest) tag efficientnet, mobilenet as xfails on vulkan (#1488) 2023-06-04 23:22:32 -07:00
Anush Elangovan
7ef1bea953 XFAIL some macos tests 2023-06-04 15:27:03 -07:00
Chi_Liu
ad89bb1413 Add distilgpt2 to stablehlo in shark tank (#1481) 2023-06-02 16:44:46 -05:00
Ean Garvey
218ed78c40 Change instances of input_type='mhlo' to 'auto' (#1482) 2023-06-02 16:43:47 -05:00
Chi_Liu
ba8eddb12f Add GPT3/OPT to Stablehlo in shark tank (#1468) 
Co-authored-by: AmosLewis <Amos_Lewsi@foxmail.com>
Co-authored-by: Ean Garvey <87458719+monorimet@users.noreply.github.com>
 2023-05-29 21:58:39 -07:00
Ean Garvey
3cddd01b10 Update OPT tokenizer and xfail a few more large tests on macos CI (#1459) 
* Update opt_torch_test.py

* Update all_models.csv
 2023-05-23 14:36:57 -07:00
Chi_Liu
64c2b2d96b Add gpt2 to stablehlo support in shark tank (#1447) 
- Add torch decomposition support when generating shark tank
- Add gpt2 stablehlo
 2023-05-22 10:45:51 -07:00
Chi_Liu
a0097a1ead Add mlir_type for torch_model_list.csv (#1428) 
- Enable stablehlo/tosa mlir output for torch model
- Add BERT stablehlo support
 2023-05-15 10:23:54 -07:00
Ean Garvey
4c07e47e8c Specify a few models for expected failure on CUDA CI. (#1430) 2023-05-12 17:03:37 -05:00
Ean Garvey
6c14847d1f xfail some large tests on macOS builder and switch to hash updates. (#1341) 
* Update test-models.yml

* Disable large tests on macOS builder
 2023-05-04 19:47:03 -05:00
Daniel Garvey
7330729c92 enable sd pytest (#1322) 2023-04-19 22:11:30 -05:00
Ean Garvey
2191fc8952 Separate pytest benchmark modes and fix model updates for SHARK downloader / pytest. (#1264) 
* Only xfail windows models in CI

* downloader: make model updates more robust.

* Separate baseline and native benchmarks in pytest.

* Fix native benchmarks

* Fix torchvision model utils.
 2023-04-03 08:24:21 -07:00
Ean Garvey
d6f740b998 allow pytest to retry getting model artifacts + disable autotuning for pytorch benchmarks (#1257) 
* Adds a few xfails to enable macOS builder

* Convert string batch sizes to ints where needed.

* allow pytest to retry getting model artifacts

* Reduce attempts and add assert msg.
 2023-03-28 23:38:45 -05:00
Ean Garvey
96b1560da5 Make batch size configurable via pytest and fix sharktank generation. (#1227) 
* Fix sharktank generation and add batch_size pytest option for torch.

* Disable torch dynamo until py3.11 supported

* Compile torchmodel without dynamo if torch.compile fails

* Use release versions of TF/Keras for importer.

* Pin torchvision and remove debug prints.

* Remove duplicates from torch model list.

* Update generate_sharktank.py

* xfail a few models that fail sharktank generation/ numerics
 2023-03-28 14:33:39 -05:00
Ean Garvey
7619e76bbd Disable and xfail some models that fail validation/compilation. (#1251) 
* Rollback T5 models for torch as the inputs give some issues that aren't trivial to resolve
* xfail efficientnet-b0 on torch+cuda -- see CUDA requesting shared memory size larger than allowed size openxla/iree#12771
 2023-03-27 12:42:53 -05:00
Daniel Garvey
9267eadbfa disable openjourney gen for nightly (#1249) 2023-03-27 11:55:34 -05:00
Boian Petkantchin
4fac46f7bb In models testing fix paths to be relative to the script dir not cwd (#1128) 
authored-by: Boian Petkantchin <boian@nod-labs.com>
 2023-03-22 15:26:52 -05:00