mirror of
https://github.com/nod-ai/SHARK-Studio.git
synced 2026-05-12 03:00:11 -04:00
-Adds date variable back to nightly.yml so shark_tank uploads are dated again -added specification for nightly pytests to not run tests on metal (vulkan is sufficient) -added some paths/filetypes to be ignored when triggering workflow runs. (no test-models on changes to .md files or anything in the shark/examples/ directory or its subdirectories. -pytest only picks up tank/test_models.py, so no need to specify which file to run when running pytest from SHARK base directory. -Cleaned up xfails so that they can be added to models as csv entries. Columns 7-9 in all_models.csv trigger xfails with cpu, cuda, vulkan, respectively, and row 10 can be populated with a reason for the xfails. -Fixed a few defaults for shark_args and pytest args (defined in conftest.py) -Fixes --update_tank option in shark_downloader removes some multiprocessing in pytest / TF+CUDA support because it breaks pytest and false passes, leaving regressions at large. -Adds xfails for and removes albert torch from gen_sharktank list (tank/torch_model_list.csv). -Cleans up xfails for cpu, cuda, vulkan (removing old ones)
3.4 KiB
3.4 KiB
| 1 | resnet50 | mhlo | tf | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | Vulkan Numerical Error: mostly conv |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | albert-base-v2 | mhlo | tf | 1e-2 | 1e-2 | default | None | False | False | False | |
| 3 | roberta-base | mhlo | tf | 1e-02 | 1e-3 | default | nhcw-nhwc | False | False | False | |
| 4 | bert-base-uncased | mhlo | tf | 1e-2 | 1e-3 | default | None | False | False | False | |
| 5 | camembert-base | mhlo | tf | 1e-2 | 1e-3 | default | None | False | False | False | |
| 6 | dbmdz/convbert-base-turkish-cased | mhlo | tf | 1e-2 | 1e-3 | default | nhcw-nhwc | True | True | True | https://github.com/iree-org/iree/issues/9971 |
| 7 | distilbert-base-uncased | mhlo | tf | 1e-2 | 1e-3 | default | None | False | False | False | |
| 8 | facebook/convnext-tiny-224 | mhlo | tf | 1e-2 | 1e-3 | tf_vit | nhcw-nhwc | True | True | True | https://github.com/nod-ai/SHARK/issues/311 & https://github.com/nod-ai/SHARK/issues/342 |
| 9 | funnel-transformer/small | mhlo | tf | 1e-2 | 1e-3 | default | None | True | True | True | https://github.com/nod-ai/SHARK/issues/201 |
| 10 | google/electra-small-discriminator | mhlo | tf | 1e-2 | 1e-3 | default | None | False | False | False | |
| 11 | google/mobilebert-uncased | mhlo | tf | 1e-2 | 1e-3 | default | None | True | False | False | Fails during iree-compile. |
| 12 | google/vit-base-patch16-224 | mhlo | tf | 1e-2 | 1e-3 | tf_vit | nhcw-nhwc | False | False | True | Vulkan Numerical Error (mostly conv) |
| 13 | microsoft/MiniLM-L12-H384-uncased | mhlo | tf | 1e-2 | 1e-3 | tf_hf | None | True | False | False | Fails during iree-compile. |
| 14 | microsoft/layoutlm-base-uncased | mhlo | tf | 1e-2 | 1e-3 | default | None | False | False | False | |
| 15 | microsoft/mpnet-base | mhlo | tf | 1e-2 | 1e-2 | default | None | False | False | False | |
| 16 | albert-base-v2 | linalg | torch | 1e-2 | 1e-3 | default | None | True | True | True | issue with aten.tanh in torch-mlir |
| 17 | alexnet | linalg | torch | 1e-2 | 1e-3 | default | None | False | False | True | Assertion Error: Zeros Output |
| 18 | bert-base-cased | linalg | torch | 1e-2 | 1e-3 | default | None | False | False | False | |
| 19 | bert-base-uncased | linalg | torch | 1e-2 | 1e-3 | default | None | False | False | False | |
| 20 | facebook/deit-small-distilled-patch16-224 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | True | False | Fails during iree-compile. |
| 21 | google/vit-base-patch16-224 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | True | False | https://github.com/nod-ai/SHARK/issues/311 |
| 22 | microsoft/beit-base-patch16-224-pt22k-ft22k | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | True | False | https://github.com/nod-ai/SHARK/issues/390 |
| 23 | microsoft/MiniLM-L12-H384-uncased | linalg | torch | 1e-2 | 1e-3 | default | None | False | False | True | |
| 24 | microsoft/resnet-50 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | Vulkan Numerical Error (mostly conv) |
| 25 | google/mobilebert-uncased | linalg | torch | 1e-2 | 1e-3 | default | None | False | False | True | https://github.com/nod-ai/SHARK/issues/344 |
| 26 | mobilenet_v3_small | linalg | torch | 1e-1 | 1e-2 | default | nhcw-nhwc | False | True | True | https://github.com/nod-ai/SHARK/issues/388 |
| 27 | nvidia/mit-b0 | linalg | torch | 1e-2 | 1e-3 | default | None | True | True | True | https://github.com/nod-ai/SHARK/issues/343 |
| 28 | resnet101 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | Vulkan Numerical Error (mostly conv) |
| 29 | resnet18 | linalg | torch | 1e-2 | 1e-3 | default | None | True | True | True | |
| 30 | resnet50 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | Vulkan Numerical Error (mostly conv) |
| 31 | squeezenet1_0 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | https://github.com/nod-ai/SHARK/issues/388 |
| 32 | wide_resnet50_2 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | Vulkan Numerical Error (mostly conv) |
| 33 | efficientnet-v2-s | mhlo | tf | 1e-02 | 1e-3 | default | nhcw-nhwc | False | False | True | https://github.com/nod-ai/SHARK/issues/575 |
| 34 | mnasnet1_0 | linalg | torch | 1e-2 | 1e-3 | default | nhcw-nhwc | False | False | True | https://github.com/nod-ai/SHARK/issues/388 |