maxmustermann/DeepLearningExamples
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
1006 commits 6 branches 0 tags 118 MiB
Commit graph

258 commits

Author SHA1 Message Date
Krzysztof Kudrynski ab85e6cc3d Updating Tacotron2_pyt (BatchNorm init fix), Resnet_tf (cosine LR), 
Transformer_pyt (bugfix)
 2019-08-13 15:01:10 +02:00
Timothy Liu ee64eb2d9b Update formatting in AMP demo notebook 2019-08-10 12:17:14 +08:00
jconwayNV 55a03b8b4e
Added newline formatting 2019-08-09 17:13:20 -07:00
jconwayNV 0ed7d8a1d9
Added TF API to summary section 
https://www.tensorflow.org/api_docs/python/tf/train/experimental/enable_mixed_precision_graph_rewrite
 2019-08-09 17:12:20 -07:00
Timothy Liu 564ddc8946 Added demo notebook for AMP (image classification) 2019-08-09 22:00:40 +08:00
Swetha Mandava 281beab09d
consistant parameter names 
consistant parameter names in run_pretraining.py and optimization.py
 2019-07-30 17:24:19 -07:00
Swetha Mandava 438c422813 moving trtis section in readme and github issue 131 2019-07-30 13:35:10 -07:00
Swetha Mandava c15d09eed6 moving trtis section in readme and github issue 131 2019-07-30 13:27:07 -07:00
Swetha Mandava 2b3b299448 moving trtis section in readme and github issue 131 2019-07-30 13:18:12 -07:00
Przemek Strzelczyk fa400a7367 Adding Jasper/PyT 2019-07-26 20:08:16 +02:00
nvpstr 71fdde7200
Merge pull request #126 from NVIDIA/nvpstr/release 
Nvpstr/release
 2019-07-25 21:06:52 +02:00
Przemek Strzelczyk a3dba86d52 Updating .gitmodules 2019-07-25 20:58:38 +02:00
nvpstr 6c0b5e36b4
Merge pull request #125 from NVIDIA/nvpstr/release 
Updating BERT with TRT-IS support and new results
 2019-07-25 20:47:32 +02:00
Przemek Strzelczyk c3ec91c6b9 Reverting changes on run_pretrainng.py 2019-07-25 20:43:32 +02:00
Przemek Strzelczyk 7a0d4fb8be Merge branch 'nvpstr/release' of https://gitlab-master.nvidia.com/dl/JoC/examples into nvpstr/release 2019-07-25 20:16:36 +02:00
Przemek Strzelczyk 15075aab0c Removing obsolete scripts 2019-07-25 20:15:52 +02:00
nvpstr 7fec94c592
Merge pull request #112 from mwawrzos/master 
[TensorFlow/Detection/SSD][issue #102] use dataset preprocessing script from tensorflow/object_detection
 2019-07-25 18:47:05 +02:00
Przemek Strzelczyk 566f9cba25 Removing some unnecessary marks 2019-07-25 07:58:23 -07:00
Przemek Strzelczyk 785801d699 Fixing Feature Support Matrix in README.md 2019-07-25 07:56:32 -07:00
Przemek Strzelczyk 8218872051 Updating BERT with TRT-IS support and new results 2019-07-25 16:53:05 +02:00
nvpstr a230e06917
Update Dockerfile 2019-07-22 16:07:25 +02:00
Marek Wawrzos c5829f0b44 use dataset preprocessing script from tensorflow/object_detection 2019-07-17 21:56:41 +02:00
nvpstr 81b1b024e1
Update Dockerfile 2019-07-16 21:57:29 +02:00
Przemek Strzelczyk fde2be5bb6 Removing flawed Transformer/TF 2019-07-16 21:30:03 +02:00
Przemek Strzelczyk a644350589 Updating models and adding BERT/PyT 
Tacotron2+Waveglow/PyT
* AMP support
* Data preprocessing for Tacotron 2 training
* Fixed dropouts on LSTMCells

SSD/PyT
* script and notebook for inference
* AMP support
* README update
* updates to examples/*

BERT/PyT
* initial release

GNMT/PyT
* Default container updated to NGC PyTorch 19.05-py3
* Mixed precision training implemented using APEX AMP
* Added inference throughput and latency results on NVIDIA Tesla V100 16G
* Added option to run inference on user-provided raw input text from command line

NCF/PyT
* Updated performance tables.
* Default container changed to PyTorch 19.06-py3.
* Caching validation negatives between runs

Transformer/PyT
* new README
* jit support added

UNet Medical/TF
* inference example scripts added
* inference benchmark measuring latency added
* TRT/TF-TRT support added
* README updated

GNMT/TF
* Performance improvements

Small updates (mostly README) for other models.
 2019-07-16 21:13:08 +02:00
Przemek Strzelczyk 0663b67c1a Updating models 2019-07-08 22:51:28 +02:00
nvpstr bf1b29d73e
Merge pull request #63 from khcs/bert/pretrain-fix 
Fix FLAG.save_checkpoint_steps naming inconsistency.
 2019-06-06 16:02:28 +02:00
Hoo Chang Shin 37c4083c4b change to BERT-base, and checkpoints-dir. 2019-05-31 10:31:26 -04:00
Hoo Chang Shin 1d8ac4b295 Merge branch 'bert/pretrain-fix' 2019-05-30 12:58:03 -04:00
Hoo Chang Shin 65bf7dbe48 initial commit on training PubMed-gvocab-base. 2019-05-30 12:55:57 -04:00
Hoo Chang Shin f3c1cc7204 Fix FLAG.save_checkpoint_steps naming inconsistency. 2019-05-30 11:37:46 -04:00
nvpstr e5a0a056d2
Update README.md 2019-05-30 13:42:31 +02:00
nvpstr b7eff3da41
Merge pull request #60 from swethmandava/smandava/bug_fix 
num_gpus pretraining fix
 2019-05-30 13:25:58 +02:00
Swetha Mandava b25181c9b8 num_gpus pretraining fix 2019-05-28 09:52:52 -07:00
nvpstr 12c7bc2302
Merge pull request #45 from eric-haibin-lin/patch-1 
Fix a few typos, and add gluonnlp implementation link
 2019-05-25 01:37:31 +02:00
Przemek Strzelczyk d2bc3da0a1 Changes in TF models: 
* added UNet for medical image segmentation
* added TF-AMP support for RN50
* small updates for other models (READMEs, benchmark & testing scripts)
 2019-05-25 01:23:11 +02:00
nvpstr 3cf8d6c3c5
Merge pull request #47 from swethmandava/master 
fix mismatch in cl arguments for BERT
 2019-05-20 13:59:49 +02:00
Swetha Mandava c5951be727
fix mismatch in cl arguments 2019-05-19 10:10:58 -07:00
Haibin Lin e46eed4c57
Update README.md 2019-05-18 22:41:42 -07:00
Khoa Ho b0d259264e
Typo/stylistic fix for README title 2019-05-14 10:30:27 -07:00
nvpstr a9d0554e5d
Merge pull request #30 from swethmandava/smandava/readme_edit 
changing perf numbers to steady state throughput
 2019-05-13 17:32:30 +02:00
Swetha Mandava 5551752822
Update readme error 2019-04-18 11:51:46 -07:00
Swetha Mandava 83fceae509 changing perf numbers to steady state throughput 2019-04-17 14:16:40 -07:00
nvpstr 51d3dceb3f
Merge pull request #24 from swethmandava/smandava/readme_edit 
fix readme
 2019-04-08 18:18:39 +02:00
Swetha Mandava ec9eaf732a removing fast_math, etc from README 2019-04-04 13:14:01 -07:00
Swetha Mandava 42c6a147f2 extra argument removed 2019-04-01 11:30:36 -07:00
Swetha Mandava bb0332437b fix readme 2019-03-28 16:10:54 -07:00
Przemek Strzelczyk 558917b3c2 Additional cleanup (BERT) 2019-03-27 17:47:46 +01:00
Przemek Strzelczyk 93fa95e3c7 Removing obsolete code (BERT) 2019-03-27 17:34:23 +01:00
Michał Krzysztof Filipiuk 041c35247b Fixed indentation 2019-03-25 12:52:08 -07:00
Swetha Mandava dfbce5d178 removing multinode scripts 2019-03-21 23:02:48 -07:00
Przemek Strzelczyk 4b09b347a7 BERT README update 2019-03-19 17:12:43 +01:00
nvpstr 21024d9126
Merge pull request #15 from NVIDIA/nvpstr/master 
Updating SSD and BERT.
 2019-03-19 16:52:05 +01:00
Jonathan DEKHTIAR f0ec1d4092
[UNet-TF Industrial] - Readme Fix 2019-03-19 16:36:25 +02:00
Przemek Strzelczyk 3d90b6dbf0 Updating SSD and BERT. 2019-03-19 15:16:07 +01:00
Przemek Strzelczyk 7a8544e0a2 Adding 9 new models: ResNet50/TF, SSD/TF, BERT/TF, NCF/TF, UNet/TF, GNMT/TF, SSD/PyT, MaskRCNN/PyT, Tacotron2+Waveglow/PyT 2019-03-18 19:45:05 +01:00
Nathan Luehr 134db41c81 Cleanup TF imagenet readme. 2018-05-03 08:47:20 -07:00
Nathan Luehr 2fdaec9c85 Add TensorFlow examples 2018-05-02 17:54:22 -07:00