pytorch/benchmark

Pull Requests Commits

Fix the broken tests caused by vision model code change

xuzhao9 committed 2 years ago

c0fc2e48

Bump up the transformers version (#1511)

desertfire committed 2 years ago

159e58f0

Add userbenchmark doc in readme (#1506)

spzala committed 2 years ago

c45fe088

Copy the nvidia libraries to the default library path (#1510)

xuzhao9 committed 2 years ago

0c4c312b

Fix wrong if check (#1507)

FindHao committed 2 years ago

0faa0142

Add newly_run and no_longer_run metrics to output yaml (#1509)

janeyx99 committed 2 years ago

443c7016

Have batch_size=1 whenever possible, add more exclusions (#1508)

janeyx99 committed 2 years ago

4b0a89a8

Fix the Inductor speedup benchmark

xuzhao9 committed 2 years ago

575b6b99

Add optim.step() benchmarks (#1473)

janeyx99 committed 2 years ago

c2ef52a6

Expose runtime errors in the stableness test (#1487)

xuzhao9 committed 2 years ago

99340dd7

Re-enable CPU device on torchrec_dlrm (#1503)

xuzhao9 committed 2 years ago

3089e825

Add typical GNN workloads (#1422)

yanbing-j committed 2 years ago

d618fa8e

Disable CPU device on torchrec_dlrm (#1501)

xuzhao9 committed 2 years ago

ed2ee259

Add regression detector to generate an abtesting config file (#1498)

xuzhao9 committed 2 years ago

28fb420b

Fix a bug in the bisection v2 workflow (#1500)

xuzhao9 committed 2 years ago

d31ce4e9

dist.init_process_group only if it's not already init'd (#1499)

janeyx99 committed 2 years ago

a6843e77

Make jit a kwarg with default value False in 4 models (#1497)

janeyx99 committed 2 years ago

8c786c25

Add necessary weasyprint dependencies to doctr_(det|reco)_predictor (#1496)

janeyx99 committed 2 years ago

dafb5c2c

Enable fx int8 for most models on cpu device (#1485)

chuanqi129 committed 2 years ago

3455e155

Update DALLE2_pytorch (#1490)

FindHao committed 2 years ago

0693b4bf

Install the latest cmake to build audio (#1495)

xuzhao9 committed 2 years ago

97cfa4d9

ddp_experiments script: add nccl-socket-ifname arg, fix timeout (#1489)

davidberard98 committed 2 years ago

4051d29c

Create the new pytorch nightly userbenchmark for V3 (#1482)

xuzhao9 committed 2 years ago

2ad17632

Improve the run.py output message

xuzhao9 committed 2 years ago

268fa024

Improve the model-stableness workflow to dump stableness result in YAML (#1483)

xuzhao9 committed 2 years ago

355dc800

Upload benchmark metrics to S3 bucket (#1484)

xuzhao9 committed 2 years ago

cc8b9f56

Enable the a10g instance ci (#1476)

xuzhao9 committed 2 years ago

83a316df

Add LLAMA (#1446)

msaroufim committed 2 years ago

c78f1f32

Replace the numpy.meshgrid() with more efficient torch.meshgrid() (#1475)

CuiJinku committed 2 years ago

d8e5325b

Fix a bug in hf_bert distributed (#1477)

xuzhao9 committed 2 years ago

0f02ca65

Newer Older