16 Best Frameworks for Deep Learning

Big Data Digest Works

Compiler: Insect-stricken, Shi Jintian, Jiang Baoshan

Deep learning is a machine learning method based on evidence learning of data, which has been developing and popular in recent years.

As a relatively new concept, for beginners who want to enter the field, or veterans who are already familiar with the method, the accessible learning resources are too rich.

In order not to be eliminated by the ever-changing technology and trends, it is a good way to actively participate in the learning and interaction of open source projects in the in-depth learning community.

In this paper, we will introduce the 16 most popular open source platforms and libraries for in-depth learning in GitHub in detail. In addition, there are some good platforms and frameworks which have not entered the list, and the abstract bacteria are listed for your reference.

GitHub has the highest collection and contribution rate of 16 open source deep learning frameworks. The greener the circle, the newer the frame is, and the bluer the color, the earlier the frame is.

TensorFlow tops the list, followed by Keras and Cafe. Below is a summary of the bacteria to share these resources to you.

16 Best Open Source Frameworks and Platforms for Deep Learning

TensorFlow

TensorFlow was originally developed by researchers and engineers at Google Brain Team in Google's Macine Intelligence Research Organization. This framework aims at facilitating researchers'research on machine learning and simplifying the migration process from research models to actual production.

Collection: 96655, contribution: 1432, procedure submission: 31714, establishment date: November 1, 2015.

Links:

Https://github.com/tensorflow/tensorflow

Keras

Keras is an API for advanced neural networks written in Python, which can be used in conjunction with TensorFlow, CNTK or Theano.

Collection: 28385, contribution: 653, procedure submission: 4468, establishment date: March 22, 2015.

Links:

Https://github.com/keras-team/keras

Caffe

Caffe is a deep learning framework that focuses on expressiveness, speed and modularity. It is developed by Berkeley Vision and Learning Center (Berkeley Visual and Learning Center) and community contributors.

Collection: 23750, contribution: 267, procedure submission: 4128, establishment date: September 8, 2015.

Links:

Https://github.com/BVLC/caffe

Microsoft Cognitive Toolkit

Microsoft Cognitive Toolkit (formerly called CNTK) is a unified tool set for in-depth learning. It describes neural networks as a series of computational steps represented by directed graphs.

Collection: 14243, contribution: 174, procedure submission: 15613, establishment date: 27 July 2014.

Links:

Https://github.com/Microsoft/CNTK

PyTorch

PyTorch is a framework of tensor computation and dynamic neural network with powerful GPU support, which is integrated with Python.

Collection: 14101, contribution: 601, procedure submission: 10733, establishment date: January 22, 2012.

Links:

Https://github.com/pytorch/pytorch

Apache MXnet

Apache MXnet is a deep learning framework designed to improve efficiency and flexibility. It allows users to mix symbolic programming with imperative programming to maximize efficiency and productivity.

Collection: 13699, contribution: 516, procedure submission: 6953, establishment date: April 26, 2015.

Links:

Https://github.com/apache/incubator-mxnet

Deep Learning 4J

Like ND4J, DataVec, Arbiter and RL4J, DeepLearning 4J is part of Skymind Intelligence Layer. It is an open source distributed neural network library written in Java and Scala and has been certified by Apache 2.0.

Collection: 8725, contribution: 141, procedure submission: 9647, establishment date: November 24, 2013.

Links:

Https://github.com/deeplearning 4j/deeplearning 4J

Theano

Theano can efficiently process user definitions, optimizations, and mathematical expressions about multidimensional arrays. But in September 2017, Theano announced that no further significant progress would be made after the 1.0 release. But don't be disappointed, Theano is still a very powerful library to support your in-depth study.

Collection: 8141, contribution: 329, procedure submission: 27974, establishment date: January 6, 2008.

Links:

Https://github.com/Theano/Theano

TFLearn

TFLearn is a modular and transparent deep learning library, which is based on TensorFlow. It aims to provide a higher level API for TensorFlow to facilitate and accelerate experimental research, and maintain complete transparency and compatibility.

Collection: 7933, contribution: 111, procedure submission: 589, establishment date: March 27, 2016.

Links:

Https://github.com/tflearn/tflearn

Torch

Torch is the main software package in Torch7, which defines the data structure and mathematical operations for multidimensional tensors. In addition, it provides many practical software for accessing files, serializing objects of any type, etc.

Collection: 7834, contribution: 133, procedure submission: 1335, establishment date: January 22, 2012.

Links:

Https://github.com/torch/torch7

Caffe2

Caffe2 is a lightweight in-depth learning framework with modularity and scalability. It improves on the basis of Caffe and improves its expressiveness, speed and modularity.

Collection: 7813, contribution: 187, procedure submission: 3678, establishment date: January 21, 2015.

Please read the Chinese version for details.