capybaralet / tensorpack

Neural Network Toolbox on TensorFlow

tensorpack

Neural Network Toolbox on TensorFlow.

See some examples to learn about the framework:

Vision:

• DoReFa-Net: train binary / low-bitwidth CNN on ImageNet
• Train ResNet on ImageNet / Cifar10 / SVHN
• InceptionV3 on ImageNet
• Fully-convolutional Network for Holistically-Nested Edge Detection(HED)
• Spatial Transformer Networks on MNIST addition
• Visualize Saliency Maps by Guided ReLU
• Similarity Learning on MNIST

Reinforcement Learning:

• Deep Q-Network(DQN) variants on Atari games, including DQN, DoubleDQN, DuelingDQN.
• Asynchronous Advantage Actor-Critic(A3C) with demos on OpenAI Gym

Unsupervised Learning:

• Generative Adversarial Network(GAN) variants, including DCGAN, InfoGAN, Conditional GAN, WGAN, DiscoGAN, Image to Image.

Speech / NLP:

• LSTM-CTC for speech recognition
• char-rnn for fun
• LSTM language model on PennTreebank

The examples are not only for demonstration of the framework -- you can train them and reproduce the results in papers.

Features:

It's Yet Another TF wrapper, but different in:

1. Not focus on models.

   • There are already too many symbolic function wrappers. Tensorpack includes only a few common models, and helpful tools such as LinearWrap to simplify large models. But you can use any other wrappers within tensorpack, such as sonnet/Keras/slim/tflearn/tensorlayer/....

2. Focus on large datasets.

   • DataFlow allows you to process large datasets such as ImageNet in Python without blocking the training.
   • DataFlow has a unified interface, so you can compose and reuse them to perform complex preprocessing.

3. Focus on training speed.

   • Tensorpack trainer is almost always faster than feed_dict based wrappers. Even on a small CNN example, the training runs 2x faster than the equivalent Keras code. More improvements to come later.

   • Data-Parallel Multi-GPU training is off-the-shelf to use. You can also define your own trainer for different style of training (e.g. GAN) without losing the efficiency.

4. Interface of extensible Callbacks. Write a callback to implement everything you want to do apart from the training iterations, and enable it with one line of code. Common examples include:

   • Change hyperparameters during training
   • Print some tensors of interest
   • Run inference on a test dataset
   • Run some operations once a while
   • Send loss to your phone

Install:

Dependencies:

• Python 2 or 3
• TensorFlow >= 1.0.0
• Python bindings for OpenCV

pip install -U git+https://github.com/ppwwyyxx/tensorpack.git
# or add `--user` to avoid system-wide installation.