mirror of https://github.com/nod-ai/AMD-SHARK-Studio.git synced 2026-04-03 03:00:17 -04:00

Files

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

opt_causallm_torch_test.py

OPT Refactor (#1516 )

2023-06-13 22:40:07 -05:00

opt_causallm.py

Pin torch-mlir and use local-task in OPT. (#1592 )

2023-06-23 19:17:05 -07:00

opt_perf_comparison_batch.py

Opt perf (#1786 )

2023-08-24 08:33:12 -05:00

opt_perf_comparison.py

Opt perf (#1786 )

2023-08-24 08:33:12 -05:00

opt_torch_test.py

Update OPT, ResNet example scripts. (#1492 )

2023-06-05 20:19:35 -07:00

README.md

Cleanup tank directory and move instructions to tank/README.md (#401 )

2022-10-13 12:20:02 -05:00

shark_hf_base_opt.py

OPT Refactor (#1516 )

2023-06-13 22:40:07 -05:00

shark_opt_wrapper_train.py

OPT Refactor (#1516 )

2023-06-13 22:40:07 -05:00

shark_opt_wrapper.py

OPT Refactor (#1516 )

2023-06-13 22:40:07 -05:00

README.md

Running Different OPT Variants

To run different sizes of OPT, change the string OPT_MODEL string in opt_torch_test.py. The default is 350m parameters. 66b cases also exist in the file, simply uncomment the test cases.