10 Most Popular Deep Learning Libraries Started in 2015

Over the past couple of days we have been posting articles on the hottest libraries created in 2015. In case you want to take a look, top 10 javascript libraries started in 2015, top 10 django libraries started in 2015, top 10 angular js libraries started in 2015.

In this post, we are going to take a look at the most popular deep learning libraries started in 2015.



Keras is a minimalist, highly modular neural networks library, written in Python and capable of running on top of either TensorFlow or Theano. It was developed with a focus on enabling fast experimentation. Being able to go from idea to result with the least possible delay is key to doing good research.


Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Go, Javascript and more. MXNet is a deep learning framework designed for both efficiency and flexibility. It allows you to mix the flavours of deep learning programs together to maximize the efficiency and your productivity.


A flexible framework of neural networks for deep learning. Chainer supports various network architectures including feed-forward nets, convnets, recurrent nets and recursive nets. It also supports per-batch architectures. Chainer supports CUDA computation. It only requires few lines of codes to leverage a GPU. It also runs on multiple GPUs with a little effort.


Deep neural network implementation without the learning cliff! This library implements multi-layer perceptrons, auto-encoders and (soon) recurrent neural networks with a stable Future Proof™ interface that's compatible with scikit-learn for a more user-friendly and Pythonic interface. It's a wrapper for powerful existing libraries such as lasagne currently, with plans for blocks.


Theano-Lights is a research framework based on Theano providing implementation of several recent Deep learning models and a convenient training and test functionality. The models are not hidden and spread out behind layers of abstraction as in most deep learning platforms to enable transparency and flexiblity during learning and research.


Highly extensible deep learning framework based on Theano.


Intel® Deep Learning Framework.

The Intel® Deep Learning Framework (IDLF) is a SDK library for Deep Neural Networks training and execution.

It includes the API that enables building neural network topology as a compute workflow, functions to optimize the graph and execute it on hardware. Our initial focus is neural network driven object classification (ImageNet topology) implemented on CPU (Xeon) and GPU (Gen).

The API is designed in the way allowing us to easily support more devices in the future. Our key principle is achieving maximum performance on each supported Intel platform.


REINFORCEjs is a Reinforcement Learning library that implements several common RL algorithms, all with web demos. In particular, the library currently includes:

Dynamic Programming methods

(Tabular) Temporal Difference Learning (SARSA/Q-Learning)

Deep Q-Learning for Q-Learning with function approximation with Neural Networks

Stochastic/Deterministic Policy Gradients and Actor Critic architectures for dealing with continuous action spaces. (very alpha, likely buggy or at the very least finicky and inconsistent)


OpenDeep is a deep learning framework for Python built from the ground up in Theano with a focus on flexibility and ease of use for both industry data scientists and cutting-edge researchers. OpenDeep is a modular and easily extensible framework for constructing any neural network architecture to solve your problem.


MXNetJS is the dmlc/mxnet Javascript package. MXNetJS brings state of art deep learning prediction API to the browser. It is generated with Emscripten and Amalgamation. MXNetJS allows you to run prediction of state-of-art deep learning models in any computational graph, and brings fun of deep learning to client side.