Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Apr 20, 2023

• [EfficientNet/TF2] remove tf async level flag

  

  Victor49152

  authored and

  nv-kkudrynski

  committed
  1e10352

  Copy full SHA for 1e10352

Commits on Apr 19, 2023

• Merge: [Jasper/PyT] Update torch.stft for PyTorch 2.0

  nv-kkudrynski

  committed
  442791a

  Copy full SHA for 442791a

• [Jasper/PyT] Update torch.stft for PyTorch 2.0

  

  alancucki

  authored and

  nv-kkudrynski

  committed
  29aaae3

  Copy full SHA for 29aaae3

Commits on Apr 18, 2023

• Added GitHub DCO workflow

  nv-kkudrynski

  authored
  c4a479b

  Copy full SHA for c4a479b

Commits on Apr 17, 2023

• [TF1] Retirement notice for all models

  nv-kkudrynski

  committed
  9de48bc

  Copy full SHA for 9de48bc

Commits on Apr 11, 2023

• [TFT/PyT] Update Checkpoints

  

  IzzyPutterman

  authored and

  nv-kkudrynski

  committed
  a6c678e

  Copy full SHA for a6c678e

Commits on Mar 31, 2023

• [TFT/PyTorch] Move to nvFuser

  

  IzzyPutterman

  authored and

  nv-kkudrynski

  committed
  777d174

  Copy full SHA for 777d174

Commits on Mar 30, 2023

• Merge: [PyT/EffDet] Replace _six.string_classes with Actual Values

  nv-kkudrynski

  committed
  e040735

  Copy full SHA for e040735

• [PyT/EffDet] Replace _six.string_classes with Actual Values

  

  suiyoubi

  authored and

  nv-kkudrynski

  committed
  0953478

  Copy full SHA for 0953478

• Merge: [BERT/Paddle] fix some issues on throughput measurement

  nv-kkudrynski

  committed
  0e2e315

  Copy full SHA for 0e2e315

• [BERT/Paddle] fix some issues on throughput measurement

  

  Wong4j

  authored and

  nv-kkudrynski

  committed
  eb4b0ab

  Copy full SHA for eb4b0ab

• Readme updates

  nv-kkudrynski

  authored
  cc14d63

  Copy full SHA for cc14d63

Commits on Mar 24, 2023

• Merge: [BERT/PyT] set default bin size to 64 for phase2

  nv-kkudrynski

  committed
  441cf4b

  Copy full SHA for 441cf4b

• [BERT/PyT] set default bin size to 64 for phase2

  

  sharathts

  authored and

  nv-kkudrynski

  committed
  a0781f8

  Copy full SHA for a0781f8

• [ConvNets/PyT] Fix interpolation type from Image.* to InterpolationMode.*

  

  AdamRajfer

  authored and

  nv-kkudrynski

  committed
  a684cf0

  Copy full SHA for a684cf0

Commits on Mar 1, 2023

• [EffDet/PyT] Invoking CUDA synchronize() before Timing

  

  suiyoubi

  authored and

  nv-kkudrynski

  committed
  f613b7c

  Copy full SHA for f613b7c

Commits on Feb 24, 2023

• Merge: [SE3Transformer/DGLPyT] Update container and fix benchmarking

  nv-kkudrynski

  committed
  acecffe

  Copy full SHA for acecffe

Commits on Feb 23, 2023

• [Maskrcnn/PyT] Synchronize before reporting DLL time

  ShriyaPalsamudram

  committed
  c44b7ad

  Copy full SHA for c44b7ad

• Readme rearrangement

  nv-kkudrynski

  committed
  6610c05

  Copy full SHA for 6610c05

• [SE3Transformer/DGLPyT] Update container and fix benchmarking

  milesial

  committed
  297e285

  Copy full SHA for 297e285

Commits on Feb 15, 2023

• Merge: [Syngen] 22.12 Release Demo notebooks description extension

  nv-kkudrynski

  committed
  678b470

  Copy full SHA for 678b470

• [Syngen] 22.12 Release Demo notebooks description extension

  

  ArturKasymov

  authored and

  nv-kkudrynski

  committed
  5c45d61

  Copy full SHA for 5c45d61

• Merge: [GNMT/PyT] Added synchronization before collecting timers, switched to correct averaging when reporting avg throughput

  nv-kkudrynski

  committed
  ed28348

  Copy full SHA for ed28348

• Merge: [TXL/PyT] Added barriers when reporting time, switched to correct averaging when reporting avg throughput

  nv-kkudrynski

  committed
  7ddd062

  Copy full SHA for 7ddd062

• Merge: [Speech models/PyT] Update perf timers and cuda syncs

  nv-kkudrynski

  committed
  959c677

  Copy full SHA for 959c677

• [Speech models/PyT] Update perf timers and cuda syncs

  

  alancucki

  authored and

  nv-kkudrynski

  committed
  5146a68

  Copy full SHA for 5146a68

• Merge: [BART/PyT] Add synchronize for benchmarking

  nv-kkudrynski

  committed
  c75cc20

  Copy full SHA for c75cc20

• Merge: [Transformer/PyTorch] Add torch.cuda.synchronize() calls

  nv-kkudrynski

  committed
  d46a356

  Copy full SHA for d46a356

• [Transformer/PyTorch] Add torch.cuda.synchronize() calls

  

  jbaczek

  authored and

  nv-kkudrynski

  committed
  efed242

  Copy full SHA for efed242

• Merge: [TFT/PyTorch] Added torch.cuda.synchronize() calls in perf meter

  nv-kkudrynski

  committed
  9868e23

  Copy full SHA for 9868e23

• [TFT/PyTorch] Added torch.cuda.synchronize() calls in perf meter

  

  jbaczek

  authored and

  nv-kkudrynski

  committed
  d190b25

  Copy full SHA for d190b25

• Merge: [DLRM/PyT] Add synchronization for more reliable benchmarking

  nv-kkudrynski

  committed
  3946f33

  Copy full SHA for 3946f33

• [DLRM/PyT] Add synchronization for more reliable benchmarking

  

  tgrel

  authored and

  nv-kkudrynski

  committed
  f5111db

  Copy full SHA for f5111db

• Merge: [nnUNet/PyT] Add torch.cuda.synchronize call while benchmarking

  nv-kkudrynski

  committed
  ff2f120

  Copy full SHA for ff2f120

Commits on Feb 13, 2023

• [BART/PyT] Add synchronize for benchmarking

  meatybobby

  committed
  a797214

  Copy full SHA for a797214

Pagination

Previous

Next