Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
634 commits
Select commit Hold shift + click to select a range
95adafd
Merge: [TXL/PyT] Include current training loss in summary generated b…
nv-kkudrynski Jun 24, 2022
43c1965
Merge: [DLRM/PyT] Log training loss as a summary metric at the end of…
nv-kkudrynski Jun 24, 2022
199e145
[ResNet50/Paddle] Adding suspend-resume functions
mingxu1067 Jun 24, 2022
55cd711
Merge: [ResNet50/Paddle] Adding suspend-resume functions
nv-kkudrynski Jun 24, 2022
e372b7b
Add logging of units to dllogger metrics
jan-golda Jun 27, 2022
a10c1d3
Merge: Add logging of units to dllogger metrics
nv-kkudrynski Jun 27, 2022
bbdd7d5
[SE3T/DGL-PyT] Log last loss as summary
milesial Jun 27, 2022
3995d2d
Merge: [SE3T/DGL-PyT] Log last loss as summary
nv-kkudrynski Jun 27, 2022
25b5392
[CI/Speech] Add val losses to converge-partial
mikolajblaz Jun 27, 2022
1f53a4d
Merge: [CI/Speech] Add val losses to converge-partial
nv-kkudrynski Jun 27, 2022
2070942
[NCF/PyT] Logging validation loss and training loss
tgrel Jun 27, 2022
d29e4f7
[GPUNet/PyT] Modified logging and added resume functionality
IzzyPutterman Jun 27, 2022
a5feffa
[ConvNets/PyT] Trition numpy dependency fixed
jkosek Jun 28, 2022
6de8b2c
[WIDENDEEP/TF1] Bump Spark from 2.3.2 to 3.1.3
nv-dmajchrowski Jun 29, 2022
71aab7b
[BERT/TF] Add final loss metrics
meatybobby Jun 30, 2022
c1584f0
[BERT/PyT] suspend-resume fixes
mwawrzos Jun 30, 2022
310ebe9
Merge: [BERT/PyT] suspend-resume fixes
nv-kkudrynski Jun 30, 2022
a53bb1d
[MaskRCNN/TF2] Fix dllogger output in train+eval mode
marta-sd Jun 30, 2022
f02611a
Merge: [MaskRCNN/TF2] Fix dllogger output in train+eval mode
nv-kkudrynski Jun 30, 2022
96928fc
[Tacotron2/PyT] Add train_loss to summary metrics
grzegorz-k-karch Jun 30, 2022
52b2215
Merge: [Tacotron2/PyT] Add train_loss to summary metrics
nv-kkudrynski Jun 30, 2022
174b3d4
Merge: [WIDENDEEP/TF1] Bump Spark from 2.3.2 to 3.1.3
nv-kkudrynski Jun 30, 2022
89c16e7
[nnUNet/PyT] Add logging for train/val loss
michal2409 Jul 4, 2022
42e1362
[nnUNet/TF2] Add logging for train/val loss
michal2409 Jul 5, 2022
e826a36
Merge: [nnUNet/PyT] Add logging for train/val loss
nv-kkudrynski Jul 5, 2022
d3ba35a
[HiFi-GAN/PyT] Remove debug code
mikolajblaz Jul 5, 2022
dbccb47
[DLRM/PyT] Add validation loss logging to test mode
tgrel Jul 6, 2022
d4d4faf
[nnUNet/TF2] Fix number epoch steps for 2D
michal2409 Jul 6, 2022
c1aa837
[EFFICIENTNET/TF2] Add a new flag to store the config information
Victor49152 Jul 7, 2022
7930dd5
Merge: [EFFICIENTNET/TF2] Add a new flag to store the config information
nv-kkudrynski Jul 7, 2022
3825249
Merge: [DLRM/PyT] Add validation loss logging to test mode
nv-kkudrynski Jul 7, 2022
c91ce88
[FastPitch/PyT] Batched CTC loss
timmoon10 Jul 7, 2022
bf4e8ab
Merge: [FastPitch/PyT] Batched CTC loss
nv-kkudrynski Jul 7, 2022
9b6381b
[ConvNets/PyT] Save N last checkpoints
asulecki Jul 8, 2022
3b05bf1
[DLRM/TF2] Port to Merlin Distributed Embeddings
tgrel Jul 11, 2022
8c7d66f
[HiFi-GAN/PyT] Set missing CUDNN API V8 env vars in CI
alancucki Jul 21, 2022
15a0678
[DLRM/TF2] Synthetic dataset bugfix
tgrel Jul 26, 2022
8bfbf82
[DLRM/PyT] Initialize the logger on the root rank only
tgrel Aug 1, 2022
d666f14
[Transformer/PyT] 22.06 release
jbaczek Aug 1, 2022
280c4f4
[NCF/PyT] Bugfixes
tgrel Aug 1, 2022
27a6bd2
Merge: [NCF/PyT] Bugfixes
nv-kkudrynski Aug 1, 2022
d7947b9
Merge: [DLRM/TF2] Synthetic dataset bugfix
nv-kkudrynski Aug 1, 2022
2c31487
Merge: [DLRM/PyT] Initialize the logger on the root rank only
nv-kkudrynski Aug 1, 2022
d549b48
[GPUNet/PyT] Add extra tests, benchmark steps to train, logging to va…
shakandrew Aug 1, 2022
d08e580
Merge: [GPUNet/PyT] Add extra tests, benchmark steps to train, loggin…
nv-kkudrynski Aug 1, 2022
39476c5
[nnUNet/PyT] Upgrade PTL to 1.6.5
michal2409 Aug 1, 2022
5b13427
Merge: [nnUNet/PyT] Upgrade PTL to 1.6.5
nv-kkudrynski Aug 1, 2022
b5977e7
[TSPP] 22.07 release
jbaczek Aug 1, 2022
dddb1d1
Merge: [TSPP] 22.07 release
nv-kkudrynski Aug 1, 2022
0e20ac8
[BERT/TF2] Add convergence log
meatybobby Aug 2, 2022
a5ca7d6
[GNMT/TF1] Fixed inference logging
mwawrzos Aug 9, 2022
38f1771
[TSPP/PyTorch] Adjust formatting and version requirements
IzzyPutterman Aug 9, 2022
d912019
Merge: [TSPP/PyTorch] Adjust formatting and version requirements
nv-kkudrynski Aug 9, 2022
d1b5b4d
[HiFi-GAN/PyT][FastPitch/PyT] Fix wrong p-arpabet defaults, support N…
alancucki Aug 9, 2022
e23e9b9
[TSPP/PyTorch] Fix log1p inverse transform
Zhylkaaa Aug 9, 2022
8909d58
Merge: [TSPP/PyTorch] Fix log1p inverse transform
nv-kkudrynski Aug 9, 2022
5e499c6
[ELECTRA/TF2] fix squad perf calculation
sharathts Aug 11, 2022
69182fc
[FastPitch/PyT] Add missing download_models.sh
alancucki Aug 12, 2022
67469e9
[FastPitch/PyT] Optimize CPU perf and remove GPU syncs
timmoon10 Aug 16, 2022
fc80de5
Merge: [FastPitch/PyT] Optimize CPU perf and remove GPU syncs
nv-kkudrynski Aug 16, 2022
87b56fc
Merge: [ELECTRA/TF2] fix squad perf calculation
nv-kkudrynski Aug 16, 2022
128ecbe
[BERT/PyT] Remove LDDL from DLE tools and point BERT to new LDDL repo
jojennin Aug 16, 2022
475cff6
Merge: [BERT/PyT] Remove LDDL from DLE tools and point BERT to new LD…
nv-kkudrynski Aug 16, 2022
077c388
[Effnet/TF2] Minor readme fix
nv-kkudrynski Aug 18, 2022
fe337c5
[SE3Transformer/DGLPyT] 22.08 container update
milesial Aug 30, 2022
206a46c
[QuartzNet/PyT][Jasper/PyT][HiFi-GAN/PyT][FastPitch/PyT] Mute pyt is_…
alancucki Sep 5, 2022
0422fa0
[Tacotron2/PyT] stabilize inference performance results for Tacotron 2
grzegorz-k-karch Sep 5, 2022
5e70f90
Merge: [Tacotron2/PyT] stabilize inference performance results for Ta…
nv-kkudrynski Sep 5, 2022
c2010f0
[Paddle/ResNet50] Adding NGC ckpt info in readme
mingxu1067 Sep 5, 2022
effe831
Merge: [Paddle/ResNet50] Adding NGC ckpt info in readme
nv-kkudrynski Sep 5, 2022
7b67399
[WideAndDeep/TF2] Add BYOD functionality
Sep 5, 2022
4d12e35
Merge: [WideAndDeep/TF2] Add BYOD functionality
nv-kkudrynski Sep 5, 2022
47d783f
[Tacotron2/PyT] Minor bug fixes
grzegorz-k-karch Sep 5, 2022
469df9b
[FastPitch/PyT][HiFi-GAN/PyT] Unidecoder: disable PyTorch warnings in…
alancucki Sep 7, 2022
69b074d
[BART/PyT] Temporary fix of pytorch lightning version
meatybobby Sep 8, 2022
937ffda
Merge: [BART/PyT] Temporary fix of pytorch lightning version
nv-kkudrynski Sep 8, 2022
84b22b3
[FastPitch/PyT] Resolve perf regression on DGX A100 + new perf tweaks
alancucki Sep 9, 2022
a98c66b
Merge: [FastPitch/PyT] Resolve perf regression on DGX A100 + new perf…
nv-kkudrynski Sep 9, 2022
63cbbdc
Merge: [SE3Transformer/DGLPyT] 22.08 container update
nv-kkudrynski Sep 9, 2022
a2f02eb
[FastPitch/PyT] Fix off-by-one in grad acc
alancucki Sep 14, 2022
f66c313
[DLRM/PyT] Fix preprocessing dockerfile
tgrel Sep 19, 2022
bdabebe
[DLRM/TF1] Explicitly install mpi4py
tgrel Sep 19, 2022
2517f61
[SE3Transformer/DGLPyT] Benchmarking fixes and tweaks
milesial Sep 19, 2022
7834973
[WideAndDeep/TF2] Add missing readme images
Sep 20, 2022
f4b9bdf
Merge: [SE3Transformer/DGLPyT] Benchmarking fixes and tweaks
nv-kkudrynski Sep 20, 2022
f47c270
Merge: [DLRM/TF1] Explicitly install mpi4py
nv-kkudrynski Sep 20, 2022
3604195
Merge: [DLRM/PyT] Fix preprocessing dockerfile
nv-kkudrynski Sep 20, 2022
6fcd5f5
[ConvNets/PyT] Fix DALI anti-alias flag
asulecki Sep 21, 2022
d21ef30
[nnUNet/TF2] Update AMP API
michal2409 Sep 15, 2022
e9f7444
[GPUNet/PyT] Add flush at end of epoch
IzzyPutterman Sep 26, 2022
971c727
[ResNet50/Paddle] Update NGC ckpt info
Wong4j Sep 27, 2022
66da9fd
[HiFi-GAN/PyT] Explicitly flush logs
alancucki Oct 3, 2022
e3b7e01
Merge: [HiFi-GAN/PyT] Explicitly flush logs
nv-kkudrynski Oct 3, 2022
5668c36
[WideAndDeep/TF2] Fix DLLogger bug
Oct 3, 2022
e976927
Merge: [WideAndDeep/TF2] Fix DLLogger bug
nv-kkudrynski Oct 3, 2022
f693f4e
Merge: [ConvNets/PyT] Fix DALI anti-alias flag
nv-kkudrynski Oct 3, 2022
5843f4e
[ConvNets/PyT] Fixed issue with unexpected keys in state_dict
asulecki Oct 3, 2022
0107cf2
Merge: [ConvNets/PyT] Fixed issue with unexpected keys in state_dict
nv-kkudrynski Oct 3, 2022
9152d5f
Merge: [nnUNet/TF2] Update AMP API
nv-kkudrynski Oct 3, 2022
2fff359
[Convents/MX] Logging and suspend-resume fixes
hXl3s Oct 3, 2022
e5efe02
Merge: [Convents/MX] Logging and suspend-resume fixes
nv-kkudrynski Oct 3, 2022
b472e61
[QuartzNet/PyT] Support NeMo checkpoints
alancucki Oct 3, 2022
87aa4b0
Merge: [QuartzNet/PyT] Support NeMo checkpoints
nv-kkudrynski Oct 3, 2022
02e1ccb
[Speech/PyT] Save two dllogs: .json and .json.X
alancucki Oct 4, 2022
60c7d97
[nnUNet/PyT] Add BraTS22 notebook
michal2409 Oct 4, 2022
72aebe7
Merge: [nnUNet/PyT] Add BraTS22 notebook
nv-kkudrynski Oct 4, 2022
d6f4301
[DLRM/TF2] Support TensorFlow 2.10
tgrel Oct 8, 2022
3408dcf
[VNet/TF1] Retirement note updated
nv-kkudrynski Oct 10, 2022
40eeae7
Merge: [VNet/TF1] Retirement note updated
nv-kkudrynski Oct 10, 2022
38079cf
Update Dockerfile
moconnor725 Oct 18, 2022
10060de
[NCF/TF1] Remove metrics that combine both performance and convergence
tgrel Oct 19, 2022
0c5e2ab
[GPUNet/PyTorch] Better resume logging functionality
IzzyPutterman Oct 24, 2022
337fae2
Merge: [GPUNet/PyTorch] Better resume logging functionality
nv-kkudrynski Oct 24, 2022
ab3a0e4
[HiFi-GAN/PyT] Remove hyphen in path: HiFi-GAN -> HiFiGAN
alancucki Oct 27, 2022
91c1de2
[HiFi-GAN/PyT] Import amp_C (apex) only when necessary
alancucki Oct 27, 2022
82b5247
[Jasper/PyT][QuartzNet/PyT] remove pyyaml requirement, as a newer ver…
mwawrzos Oct 28, 2022
9878fb8
Merge: [Jasper/PyT][QuartzNet/PyT] remove pyyaml requirement, as a ne…
nv-kkudrynski Oct 28, 2022
4e9bae8
[MRCNN/PYT] Fix coco data download
ShriyaRishab Nov 2, 2022
448be1b
[Bert/PyT] Replace conda install of jemalloc with apt package manager
jojennin Nov 3, 2022
359ec71
[nnUNet/TF2] Log best metric at the end of training
michal2409 Nov 4, 2022
afc691c
Merge: [MRCNN/PYT] Fix coco data download
nv-kkudrynski Nov 4, 2022
9e54279
Merge: [nnUNet/TF2] Log best metric at the end of training
nv-kkudrynski Nov 8, 2022
b1fc3c4
[SSD/PyT] Update Dockerfile
nv-kkudrynski Nov 10, 2022
cfdbf4e
[BERT/Paddle] BERT Paddle initial release
Wong4j Nov 20, 2022
12f50df
[EfficientDet/TF2] Add validation accuracy dllogger logging
mwawrzos Nov 22, 2022
c903326
Merge: [EfficientDet/TF2] Add validation accuracy dllogger logging
nv-kkudrynski Nov 22, 2022
bf00fe1
[SIM/TF2] Release new version of SIM model with prebatching support
tomsiadev Nov 28, 2022
4da7bf2
[nnUNet/PyT] Update container to 22.11, channel last conv, nvFuser In…
michal2409 Dec 5, 2022
09f9fe6
[nnUNet/TF2] Update container to 22.11, fix XLA+channel last conv, mu…
michal2409 Dec 5, 2022
eb35710
Merge: [nnUNet/TF2] Update container to 22.11, fix XLA+channel last c…
nv-kkudrynski Dec 5, 2022
1bd98ac
[RN50/MXNet] Release 22.10
mmarcinkiewicz Dec 13, 2022
cbbcc0f
[ConvNets/EffNetV2/TF2] model resume fixed
ntajbakhsh Dec 13, 2022
ef0128e
Merge: [ConvNets/EffNetV2/TF2] model resume fixed
nv-kkudrynski Dec 13, 2022
0e19983
[EfficientNet/TF2] Fix broken links in the ReadMe
ntajbakhsh Dec 13, 2022
9615805
Merge: [EfficientNet/TF2] Fix broken links in the ReadMe
nv-kkudrynski Dec 13, 2022
56ce6c8
[HiFi-GAN/PyT] Whitelist data/cmudict in .dockerignore
alancucki Dec 13, 2022
9c34e35
Merge: [HiFi-GAN/PyT] Whitelist data/cmudict in .dockerignore
nv-kkudrynski Dec 13, 2022
41d96d8
[nnUNet/TF2] Update requirements.txt
michal2409 Dec 14, 2022
0c3c237
[ConvNets/EffNetV2][PyT] Fix grad scaling and dropout
AdamRajfer Dec 16, 2022
4954233
Merge branch 'arajfer/fix-convergence-in-convnets' into 'internal/main'
jan-golda Dec 16, 2022
c531cd9
[ConvNets/PyT] Enable logging gradient scale
AdamRajfer Dec 16, 2022
bd1fb86
Merge: [ConvNets/PyT] Enable logging gradient scale
nv-kkudrynski Dec 16, 2022
ca5ae20
Merge: [nnUNet/TF2] Update requirements.txt
nv-kkudrynski Dec 16, 2022
c1bf9aa
[MoFlow/PyT] Initial release
marta-sd Jan 10, 2023
1dbad27
[GPUNet/PyTorch] Fix GPUNet-0 accuracy typo
IzzyPutterman Jan 10, 2023
e9ce762
Merge: [GPUNet/PyTorch] Fix GPUNet-0 accuracy typo
nv-kkudrynski Jan 10, 2023
ab3e924
[Jasper/PyT][QuartzNet/PyT] Fix Docker build, drop unused pycuda
alancucki Jan 10, 2023
64f9ecb
Merge: [Jasper/PyT][QuartzNet/PyT] Fix Docker build, drop unused pycuda
nv-kkudrynski Jan 10, 2023
cbeb6b9
[ELECTRA/TF2] Fixed build issue of tokenizer on arm
meatybobby Jan 10, 2023
4b946f4
Merge: [ELECTRA/TF2] Fixed build issue of tokenizer on arm
nv-kkudrynski Jan 10, 2023
72a15ee
[FastPitch/PyT] Bump container to 22.08, update perf results
alancucki Jan 10, 2023
84be38e
Merge: [FastPitch/PyT] Bump container to 22.08, update perf results
nv-kkudrynski Jan 10, 2023
c2bb3fe
[FastPitch/PyT] Add mixed English and Mandarin bilingual support
MuyangDu Jan 10, 2023
35d8759
[wav2vec2/PyT] Initial release
alancucki Jan 13, 2023
77b6eab
[NCF/PyT] Fix time measurement bugs
tgrel Jan 19, 2023
7ee81a1
[BERT/TF2] Change throughput to sequences per second in REAMDE
meatybobby Jan 25, 2023
a0e9004
[MoFlow/PyT] Update README
marta-sd Jan 27, 2023
0522471
Merge: [BERT/TF2] Change throughput to sequences per second in REAMDE
nv-kkudrynski Jan 27, 2023
0915477
Merge: [NCF/PyT] Fix time measurement bugs
nv-kkudrynski Jan 27, 2023
20bda77
[nnUNet/PyT] Fix case with checkpoint path set to None
michal2409 Jan 27, 2023
35feabc
Merge: [nnUNet/PyT] Fix case with checkpoint path set to None
nv-kkudrynski Jan 27, 2023
0409902
[SSD/PyT] New release with 22.10 base image
ahmadki Jan 30, 2023
305744e
[BART/PyT] Add pretraining feature support
meatybobby Jan 30, 2023
75052a4
Merge: [BART/PyT] Add pretraining feature support
nv-kkudrynski Jan 30, 2023
e6244d7
[Syngen] 22.12 Release
sajaddarabi Jan 30, 2023
0c4310b
[ResNet/PyT] Fix Resnet BasicBlock constructor
Aug 22, 2022
ef5241e
[BERT/PyT] fix for issue 1249
sharathts Jan 31, 2023
a23be21
[SE3Transformer/DGLPyT] Move results to workdir
milesial Jan 31, 2023
2586ee3
Merge: [SE3Transformer/DGLPyT] Move results to workdir
nv-kkudrynski Jan 31, 2023
def4658
Merge: [ResNet/PyT] Fix Resnet BasicBlock constructor
nv-kkudrynski Feb 1, 2023
3b4ed5d
[Speech models] Update main README model summary
alancucki Feb 1, 2023
260c8f6
Merge: [Speech models] Update main README model summary
nv-kkudrynski Feb 1, 2023
7475648
[BERT/TF1] Fix Triton build issues for ARM
meatybobby Feb 6, 2023
a07a807
[TFT/PyTorch] Adding TFT to Torchhub
IzzyPutterman Feb 6, 2023
17c268f
[SIM/TF2] Fix concat bug from TensorFlow 2.11
tomsiadev Feb 7, 2023
dff4935
Merge: [SIM/TF2] Fix concat bug from TensorFlow 2.11
nv-kkudrynski Feb 7, 2023
b9178e8
[BERT/Paddle] Update LDDL
Wong4j Feb 7, 2023
8cdaba1
Merge: [BERT/Paddle] Update LDDL
nv-kkudrynski Feb 7, 2023
0e1c6c5
[WaveGlow/PyT] Enable TorchScript
alancucki Feb 7, 2023
6a16011
Merge: [WaveGlow/PyT] Enable TorchScript
nv-kkudrynski Feb 7, 2023
9970904
[nnUNet/PyT] Add torch.cuda.synchronize call while benchmarking
michal2409 Feb 8, 2023
afea561
[TTS/Torchhub] Expose HiFiGAN and FastPitch via TorchHub
nv-kkudrynski Feb 13, 2023
327898a
[GNMT/PyT] Added synchronization before collecting timers, switched t…
szmigacz Feb 13, 2023
8f82237
[TXL/PyT] Added barriers when reporting time, switched to correct ave…
szmigacz Feb 13, 2023
a797214
[BART/PyT] Add synchronize for benchmarking
meatybobby Feb 13, 2023
ff2f120
Merge: [nnUNet/PyT] Add torch.cuda.synchronize call while benchmarking
nv-kkudrynski Feb 15, 2023
f5111db
[DLRM/PyT] Add synchronization for more reliable benchmarking
tgrel Feb 15, 2023
3946f33
Merge: [DLRM/PyT] Add synchronization for more reliable benchmarking
nv-kkudrynski Feb 15, 2023
d190b25
[TFT/PyTorch] Added torch.cuda.synchronize() calls in perf meter
jbaczek Feb 15, 2023
9868e23
Merge: [TFT/PyTorch] Added torch.cuda.synchronize() calls in perf meter
nv-kkudrynski Feb 15, 2023
efed242
[Transformer/PyTorch] Add torch.cuda.synchronize() calls
jbaczek Feb 15, 2023
d46a356
Merge: [Transformer/PyTorch] Add torch.cuda.synchronize() calls
nv-kkudrynski Feb 15, 2023
c75cc20
Merge: [BART/PyT] Add synchronize for benchmarking
nv-kkudrynski Feb 15, 2023
5146a68
[Speech models/PyT] Update perf timers and cuda syncs
alancucki Feb 15, 2023
959c677
Merge: [Speech models/PyT] Update perf timers and cuda syncs
nv-kkudrynski Feb 15, 2023
7ddd062
Merge: [TXL/PyT] Added barriers when reporting time, switched to corr…
nv-kkudrynski Feb 15, 2023
ed28348
Merge: [GNMT/PyT] Added synchronization before collecting timers, swi…
nv-kkudrynski Feb 15, 2023
5c45d61
[Syngen] 22.12 Release Demo notebooks description extension
ArturKasymov Feb 15, 2023
678b470
Merge: [Syngen] 22.12 Release Demo notebooks description extension
nv-kkudrynski Feb 15, 2023
297e285
[SE3Transformer/DGLPyT] Update container and fix benchmarking
milesial Feb 23, 2023
6610c05
Readme rearrangement
nv-kkudrynski Feb 23, 2023
c44b7ad
[Maskrcnn/PyT] Synchronize before reporting DLL time
ShriyaRishab Feb 23, 2023
acecffe
Merge: [SE3Transformer/DGLPyT] Update container and fix benchmarking
nv-kkudrynski Feb 24, 2023
f613b7c
[EffDet/PyT] Invoking CUDA synchronize() before Timing
suiyoubi Mar 1, 2023
a684cf0
[ConvNets/PyT] Fix interpolation type from Image.* to InterpolationMo…
AdamRajfer Mar 24, 2023
a0781f8
[BERT/PyT] set default bin size to 64 for phase2
sharathts Mar 24, 2023
441cf4b
Merge: [BERT/PyT] set default bin size to 64 for phase2
nv-kkudrynski Mar 24, 2023
cc14d63
Readme updates
nv-kkudrynski Mar 30, 2023
eb4b0ab
[BERT/Paddle] fix some issues on throughput measurement
Wong4j Mar 30, 2023
0e2e315
Merge: [BERT/Paddle] fix some issues on throughput measurement
nv-kkudrynski Mar 30, 2023
0953478
[PyT/EffDet] Replace _six.string_classes with Actual Values
suiyoubi Mar 30, 2023
e040735
Merge: [PyT/EffDet] Replace _six.string_classes with Actual Values
nv-kkudrynski Mar 30, 2023
777d174
[TFT/PyTorch] Move to nvFuser
IzzyPutterman Mar 31, 2023
a6c678e
[TFT/PyT] Update Checkpoints
IzzyPutterman Apr 11, 2023
9de48bc
[TF1] Retirement notice for all models
nv-kkudrynski Apr 17, 2023
c4a479b
Added GitHub DCO workflow
nv-kkudrynski Apr 18, 2023
29aaae3
[Jasper/PyT] Update torch.stft for PyTorch 2.0
alancucki Apr 19, 2023
442791a
Merge: [Jasper/PyT] Update torch.stft for PyTorch 2.0
nv-kkudrynski Apr 19, 2023
1e10352
[EfficientNet/TF2] remove tf async level flag
Victor49152 Apr 20, 2023
5bc69ca
[DLRM/PyT] Stop using apex AMP and DDP
tgrel Apr 24, 2023
05ee986
[NCF/PyT] Stop using deprecated apex AMP and apex DDP
tgrel Apr 24, 2023
f81fca9
[ResNet50/Paddle] Do inference with synthetic input as default
leo0519 May 8, 2023
d56fe70
[UNet3+/TF2] Initial contribution (#1267)
hamidriasat May 8, 2023
7b89aed
[Transformer/PyT] minor bugfix
jbaczek May 15, 2023
9becdf8
[Jasper/PyT, QuartzNet/PyT] Update Pandas and Dali versions
alancucki May 22, 2023
370a221
Merge: [DLRM/PyT] Stop using apex AMP and DDP
nv-kkudrynski May 29, 2023
2a7c251
Merge: [NCF/PyT] Stop using deprecated apex AMP and apex DDP
nv-kkudrynski May 29, 2023
810bcf3
[resnet/mxnet] Apply horovod patch for hvd init
mmarcinkiewicz May 29, 2023
54e2fb4
Merge: [resnet/mxnet] Apply horovod patch for hvd init
nv-kkudrynski May 29, 2023
8ed53a4
[Jasper/PyT, QuartzNet/PyT] Fix Ada L40 on 23.06 base container
alancucki Jun 30, 2023
d53419f
[DLRM/TF2] DLRM and DCNv2 23.02 release
tgrel Jun 30, 2023
2693c63
Merge: [DLRM/TF2] DLRM 23.02 release
nv-kkudrynski Jun 30, 2023
fc9c09b
[ResNet/Paddle] Add CUDNNv8 ResUnit fusion
Tom-Zheng Jul 26, 2023
820b6dd
[Efficientnet/TF2] fix keras imports
Victor49152 Jul 27, 2023
96bdb5b
Merge: [Efficientnet/TF2] fix keras imports
nv-kkudrynski Jul 27, 2023
296bb99
[SynGen] 23.08 Release
ArturKasymov Aug 4, 2023
a5388a4
[BERT/Paddle] Update base image and integrate cuDNN fused MHA
Wong4j Aug 23, 2023
41f582b
[JAX] Add JAX models with reference to Rosetta Github
sharathts Sep 5, 2023
da7e1a7
[DLRM/TF2] CPU offloading
tgrel Oct 3, 2023
6f3a71a
[JAX/Imagen] Imagen model with reference to Rosetta Github
sharathts Oct 6, 2023
e36f9d9
[DLRM/TF2] Fix numpy bool API change
tgrel Nov 13, 2023
e52bcb0
[DLRM/PyT] Fix np.bool API deprecation
tgrel Nov 13, 2023
b849275
Merge: [DLRM/PyT] Fix np.bool API deprecation
nv-kkudrynski Nov 30, 2023
34770bb
[Jasper/PyT,QuartzNet/PyT] use the dafault DALI installation from the…
mwawrzos Dec 4, 2023
0131db6
Merge branch 'mwawrzos/nvbug/4393747' into 'internal/main'
nv-kkudrynski Dec 4, 2023
4fca54f
[RN50/Paddle] Fix 2308 compatibility issue
Wong4j Dec 7, 2023
9dd9fcb
[wav2vec2.0/PyT] Fix pip dependencies (librosa - numpy)
alancucki Dec 8, 2023
38934f9
[RN50/Paddle] Remove export script and add INT8 feature (QAT + infere…
leo0519 Feb 20, 2024
2788e44
[UNET2D/TF2] Fix numpy API deprecation
mmarcinkiewicz Mar 11, 2024
729963d
[TSPP] 24.03 Release
nv-dmajchrowski Apr 4, 2024
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
29 changes: 29 additions & 0 deletions .github/workflows/cla.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
name: "DCO Assistant"
on:
issue_comment:
types: [created]
pull_request_target:
types: [opened,closed,synchronize]

permissions:
actions: write
contents: write
pull-requests: write
statuses: write

jobs:
DCOAssistant:
runs-on: ubuntu-latest
steps:
- name: "DCO Assistant"
if: (github.event.comment.body == 'recheck' || github.event.comment.body == 'I have read the DCO Document and I hereby sign the DCO') || github.event_name == 'pull_request_target'
uses: contributor-assistant/github-action@v2.3.0
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
with:
path-to-signatures: '.github/dco/signatures.json'
path-to-document: 'https://developercertificate.org/'
branch: 'dco-do-not-remove'
allowlist: user1,bot*
use-dco-flag: true
custom-notsigned-prcomment: '<br/>Thank you for your submission. Before we can accept your contribution, please sign our [Developer Certificate of Origin](https://developercertificate.org) by posting a comment with the content exactly as below.<br/>'
2 changes: 2 additions & 0 deletions CUDA-Optimized/FastSpeech/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -315,6 +315,8 @@ Sample result waveforms are [FP32](fastspeech/trt/samples) and [FP16](fastspeech

## Performance

The performance measurements in this document were conducted at the time of publication and may not reflect the performance achieved from NVIDIA’s latest software release. For the most up-to-date performance measurements, go to [NVIDIA Data Center Deep Learning Product Performance](https://developer.nvidia.com/deep-learning-performance-training-inference).

### Benchmarking

The following section shows how to run benchmarks measuring the model performance in training and inference modes.
Expand Down
123 changes: 123 additions & 0 deletions DGLPyTorch/DrugDiscovery/SE3Transformer/.dockerignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,123 @@
.Trash-0
.git
data/
.DS_Store
*wandb/
*.pt
*.swp

# added by FAFU
.idea/
cache/
downloaded/
*.lprof

# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class

# C extensions
*.so

# Distribution / packaging
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST

# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec

# Installer logs
pip-log.txt
pip-delete-this-directory.txt

# Unit test / coverage reports
htmlcov/
.tox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
.hypothesis/
.pytest_cache/

# Translations
*.mo
*.pot

# Django stuff:
*.log
local_settings.py
db.sqlite3

# Flask stuff:
instance/
.webassets-cache

# Scrapy stuff:
.scrapy

# Sphinx documentation
docs/_build/

# PyBuilder
target/

# Jupyter Notebook
.ipynb_checkpoints

# pyenv
.python-version

# celery beat schedule file
celerybeat-schedule

# SageMath parsed files
*.sage.py

# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/

# Spyder project settings
.spyderproject
.spyproject

# Rope project settings
.ropeproject

# mkdocs documentation
/site

# mypy
.mypy_cache/

**/benchmark
**/results
*.pkl
*.log
121 changes: 121 additions & 0 deletions DGLPyTorch/DrugDiscovery/SE3Transformer/.gitignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,121 @@
data/
.DS_Store
*wandb/
*.pt
*.swp

# added by FAFU
.idea/
cache/
downloaded/
*.lprof

# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class

# C extensions
*.so

# Distribution / packaging
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST

# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec

# Installer logs
pip-log.txt
pip-delete-this-directory.txt

# Unit test / coverage reports
htmlcov/
.tox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
.hypothesis/
.pytest_cache/

# Translations
*.mo
*.pot

# Django stuff:
*.log
local_settings.py
db.sqlite3

# Flask stuff:
instance/
.webassets-cache

# Scrapy stuff:
.scrapy

# Sphinx documentation
docs/_build/

# PyBuilder
target/

# Jupyter Notebook
.ipynb_checkpoints

# pyenv
.python-version

# celery beat schedule file
celerybeat-schedule

# SageMath parsed files
*.sage.py

# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/

# Spyder project settings
.spyderproject
.spyproject

# Rope project settings
.ropeproject

# mkdocs documentation
/site

# mypy
.mypy_cache/

**/benchmark
**/results
*.pkl
*.log
67 changes: 67 additions & 0 deletions DGLPyTorch/DrugDiscovery/SE3Transformer/Dockerfile
Original file line number Diff line number Diff line change
@@ -0,0 +1,67 @@
# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
#
# SPDX-FileCopyrightText: Copyright (c) 2021 NVIDIA CORPORATION & AFFILIATES
# SPDX-License-Identifier: MIT

# run docker daemon with --default-runtime=nvidia for GPU detection during build
# multistage build for DGL with CUDA and FP16

ARG FROM_IMAGE_NAME=nvcr.io/nvidia/pytorch:23.01-py3

FROM ${FROM_IMAGE_NAME} AS dgl_builder

ENV DEBIAN_FRONTEND=noninteractive
RUN apt-get update \
&& apt-get install -y git build-essential python3-dev make cmake \
&& rm -rf /var/lib/apt/lists/*
WORKDIR /dgl
RUN git clone --branch 1.0.0 --recurse-submodules --depth 1 https://github.com/dmlc/dgl.git .
WORKDIR build
RUN export NCCL_ROOT=/usr \
&& cmake .. -GNinja -DCMAKE_BUILD_TYPE=Release \
-DUSE_CUDA=ON -DCUDA_ARCH_BIN="60 70 80" -DCUDA_ARCH_PTX="80" \
-DCUDA_ARCH_NAME="Manual" \
-DUSE_FP16=ON \
-DBUILD_TORCH=ON \
-DUSE_NCCL=ON \
-DUSE_SYSTEM_NCCL=ON \
-DBUILD_WITH_SHARED_NCCL=ON \
-DUSE_AVX=ON \
&& cmake --build .


FROM ${FROM_IMAGE_NAME}

WORKDIR /workspace/se3-transformer

# copy built DGL and install it
COPY --from=dgl_builder /dgl ./dgl
RUN cd dgl/python && python setup.py install && cd ../.. && rm -rf dgl

ADD requirements.txt .
RUN pip install --no-cache-dir --upgrade --pre pip
RUN pip install --no-cache-dir -r requirements.txt
ADD . .

ENV DGLBACKEND=pytorch
ENV OMP_NUM_THREADS=1


7 changes: 7 additions & 0 deletions DGLPyTorch/DrugDiscovery/SE3Transformer/LICENSE
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
Copyright 2021-2022 NVIDIA CORPORATION & AFFILIATES

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
8 changes: 8 additions & 0 deletions DGLPyTorch/DrugDiscovery/SE3Transformer/NOTICE
Original file line number Diff line number Diff line change
@@ -0,0 +1,8 @@

SE(3)-Transformer PyTorch

This repository includes software from https://github.com/FabianFuchsML/se3-transformer-public
licensed under the MIT License.

This repository includes software from https://github.com/lucidrains/se3-transformer-pytorch
licensed under the MIT License.
Loading