pytorch/benchmark

Pull Requests Commits

address wills feedback

Krovatkin committed 3 years ago

90ad9713

WIP changes to Background_Matting that seem necessary but not sufficient to get it running on lazy

wconstab committed 3 years ago

367db6e4

Move the computation of self.example_outputs to be training only

desertfire committed 3 years ago

ba9d1a22

Improve the `run_sweep.py` script to support precision attribute (#844)

xuzhao9 committed 3 years ago

356fd5da

Add accelerate to requirements.txt (#849)

mrshenli committed 3 years ago

86d29d08

Upgrade to cudnn 8.3.2 (#847)

xuzhao9 committed 3 years ago

ff711465

Pin cmake version because recent cmake upgrade breaks pytorch (#845)

xuzhao9 committed 3 years ago

f1c00d01

Fix 'RecursiveScriptModule' object has no attribute 'forward' in demucs (#564)

jansel committed 3 years ago

ce5186d3

Fix `get_module()` bug in timm_efficientdet (#842)

xuzhao9 committed 3 years ago

b1e7fe8b

Remove print from speech_transformer (#837)

jansel committed 3 years ago

0c840c14

Disable moco on CPU (#838)

jansel committed 3 years ago

64268fac

Cache the huggingface model at install stage (#836)

xuzhao9 committed 3 years ago

332055b4

Add initial fambench support to e2e_models (#821)

xuzhao9 committed 3 years ago

4809ecf6

Refactor the huggingface model code (#825)

xuzhao9 committed 3 years ago

cf34503e

Adding the model input iterators. (#806)

xuzhao9 committed 3 years ago

c0c09d4d

Add the efficientdet model (#582)

xuzhao9 committed 3 years ago

38e31bdc

Fix GPU memory leak introduced by correctness checking (#830)

xuzhao9 committed 3 years ago

03372976

fix lower_to_trt api (#822)

Wei Wei committed 3 years ago

21cfa45f

Don't load empty directory in unit tests (#823)

xuzhao9 committed 3 years ago

f8ab5657

Improve the microbenchmark framework (#819)

xuzhao9 committed 3 years ago

1aea85ab

Add TorchDynamo support (#816)

xuzhao9 committed 3 years ago

45db3b07

attention issue fix for fp16 inference (#794)

Wei Wei committed 3 years ago

d56a6259

Remove numba deps (#811)

xuzhao9 committed 3 years ago

aabd309f

NVFuser microbenchmarks and filtering tool (#793)

davidberard98 committed 3 years ago

67d08eac

Revert D34929680: Multisect successfully blamed D34929680 for test failures (#74381)

Dark Knight committed 3 years ago

5b38fc02

Disable correctness ci because it is flaky (#807)

xuzhao9 committed 3 years ago

cad410ef

Replacing fp16 and int8 mode with enum type (#74338)

Mengchi Zhang committed 3 years ago

b5d4646c

Enable fp16 half on timm models (#802)

xuzhao9 committed 3 years ago

c6df90c1

Checkout the main branch before finding the correct commit on date. (#804)

xuzhao9 committed 3 years ago

bff77519

Clean up the leaked files in the root directory (#792)

xuzhao9 committed 3 years ago

d39385de

Older