Skip to content
Branch: master
Find file History
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
A3C-Gym Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
CTC-TIMIT docs cleanup Sep 2, 2019
CaffeModels Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
Char-RNN Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
ConvolutionalPoseMachines Move Caffe models together. Mar 8, 2018
DeepQNetwork docs cleanup Sep 2, 2019
DisturbLabel Rename PrefetchData -> MultiProcessRunner May 25, 2019
DoReFa-Net Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
DynamicFilterNetwork Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
FasterRCNN [MaskRCNN] fallback to tf.gather_nd for older TF versions Sep 20, 2019
GAN docs cleanup Sep 2, 2019
HED Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
ImageNetModels Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
OpticalFlow Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
PennTreebank Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
ResNet Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
Saliency Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
ShuffleNet Move ImageNet models together Mar 8, 2018
SimilarityLearning Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
SpatialTransformer Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
SuperResolution Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
basics Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
keras TowerFuncWrapper -> TowerFunc Aug 22, 2019
README.md update docs Aug 1, 2019
boilerplate.py Use SmartInit globally - a simpler interface to initialization Sep 1, 2019
tox.ini linter actions Aug 30, 2019

README.md

Tensorpack Examples

Training examples with reproducible performance.

The word "reproduce" should always mean reproduce performance. With the magic of SGD, wrong deep learning code often appears to work, especially if you try it on toy datasets. Github is full of deep learning code that "implements" but does not "reproduce" methods, and you'll not know whether the implementation is actually correct. See Unawareness of Deep Learning Mistakes.

We refuse toy examples. Instead of showing tiny CNNs trained on MNIST/Cifar10, we provide training scripts that reproduce well-known papers.

We refuse low-quality implementations. Unlike most open source repos which only implement methods, Tensorpack examples faithfully reproduce experiments and performance in the paper, so you're confident that they are correct.

Getting Started:

These are the only toy examples in tensorpack. They are supposed to be just demos.

Vision:

Name Performance
Train ResNet, ShuffleNet and other models on ImageNet reproduce 10 papers
Train Mask/Faster R-CNN on COCO reproduce 7 papers
Generative Adversarial Network(GAN) variants, including DCGAN, InfoGAN,
Conditional GAN, WGAN, BEGAN, DiscoGAN, Image to Image, CycleGAN
visually reproduce 8 papers
DoReFa-Net: training binary / low-bitwidth CNN on ImageNet reproduce 4 papers
Adversarial training with state-of-the-art robustness official code for the paper
Fully-convolutional Network for Holistically-Nested Edge Detection(HED) visually reproduce
Spatial Transformer Networks on MNIST addition reproduce the paper
Visualize CNN saliency maps visually reproduce
Similarity learning on MNIST
Single-image super-resolution using EnhanceNet
Learn steering filters with Dynamic Filter Networks visually reproduce
Load a pre-trained AlexNet, VGG, or Convolutional Pose Machines
Load a pre-trained FlowNet2-S, FlowNet2-C, FlowNet2

Reinforcement Learning:

Name Performance
Deep Q-Network(DQN) variants on Atari games, including
DQN, DoubleDQN, DuelingDQN.
reproduce the paper
Asynchronous Advantage Actor-Critic(A3C) on Atari games reproduce the paper

Speech / NLP:

Name Performance
LSTM-CTC for speech recognition reproduce the paper
char-rnn for fun fun
LSTM language model on PennTreebank reproduce reference code
You can’t perform that action at this time.