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

24 commits

Author SHA1 Message Date
gkarch b8027d8914 added jupyter notebook 2019-09-03 18:40:29 +02:00
Krzysztof Kudrynski ab85e6cc3d Updating Tacotron2_pyt (BatchNorm init fix), Resnet_tf (cosine LR), 
Transformer_pyt (bugfix)
 2019-08-13 15:01:10 +02:00
gkarch b56f72bcd9 fixed audio 2019-07-30 15:01:15 +02:00
Grzegorz Karch d0c6294695 removed cudnn-benchmark from tacotron2 perf command lines in the readme 2019-07-24 01:08:20 -07:00
Grzegorz Karch 7e5013fff9 fixed script names in the readme 2019-07-24 00:30:07 -07:00
Grzegorz Karch bb7a4ac630 updated container version in the readme, changed order of benchmarking scripts 2019-07-23 15:50:02 -07:00
Grzegorz Karch 6c42c20948 updated readme, number of epochs in training scripts 2019-07-23 15:43:24 -07:00
Grzegorz Karch 87accc3073 updated performance benchmark command lines in the readme 2019-07-23 15:26:54 -07:00
Grzegorz Karch 25b3f2678a fixed inference results 2019-07-23 14:29:58 -07:00
Grzegorz Karch 979e291848 updated to 19.07 version 2019-07-23 12:45:37 -07:00
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 f89dcca19d
Merge pull request #89 from GrzegorzKarchNV/readme-epochs 
changed number of epochs in train scripts; removed number of epochs f…
 2019-07-01 11:44:01 +02:00
Grzegorz Karch d7c20e5bfc fixed batches for mixed precision and fp32, added info when no input is provided for inference, added units to expected training time 2019-06-27 01:48:46 -07:00
Grzegorz Karch 7eb5eece14 changed number of epochs in train scripts; removed number of epochs from readme for the audio samples 2019-06-26 08:57:23 -07:00
Grzegorz Karch cd284a62ec changed text to phrase as input to inference in the readme 2019-06-07 17:37:30 +02:00
Grzegorz Karch dc7cd1ae5b updated readme 2019-06-07 17:37:30 +02:00
gkarch 79b9a48bf1 added warmup parameter to tacotron2 inference; updated README 2019-06-07 17:37:30 +02:00
gkarch 2f34442146 update perf numbers 2019-06-06 15:27:21 +02:00
gkarch 4dd9ecfcfa updated readme - warmup for inference 2019-06-06 14:16:51 +02:00
GrzegorzKarchNV 867ab68fad
Update models.py 2019-06-04 15:44:05 +02:00
Krzysztof Kudrynski 14fe91fad5 added missing nested-modules-discovery from global entrypoint like torch.hub 2019-05-21 13:32:36 +02:00
Krzysztof Kudrynski 49acab486b torch.hub entrypoints for tacotron2, waveglow and ncf 2019-05-15 14:16:46 +02:00
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