Comparison of deep learning software

The following table compares some of the most popular software frameworks, libraries and computer programs for deep learning.

Deep learning software by name

Software	Creator	Software license^[a]	Open source	Platform	Written in	Interface	OpenMP support	OpenCL support	CUDA support	Automatic differentiation^[1]	Has pretrained models	Recurrent nets	Convolutional nets	RBM/DBNs	Parallel execution (multi node)
Apache Singa	Apache Incubator	Apache 2.0	Yes	Linux, Mac OS X, Windows	C++	Python, C++, Java	No	Yes	Yes	?	Yes	Yes	Yes	Yes	Yes
Caffe	Berkeley Vision and Learning Center	BSD license	Yes	Linux, Mac OS X, Windows^[2]	C++	Python, MATLAB	Yes	Under development^[3]	Yes	Yes	Yes^[4]	Yes	Yes	No	?
Deeplearning4j	Skymind engineering team; Deeplearning4j community; originally Adam Gibson	Apache 2.0	Yes	Linux, Mac OS X, Windows, Android (Cross-platform)	java	Java, Scala, Clojure, Python (Keras)	Yes	On roadmap^[5]	Yes^[6]	Computational Graph	Yes^[7]	Yes	Yes	Yes	Yes^[8]
Dlib	Davis King	Boost Software License	Yes	Cross-Platform	C++	C++	Yes	No	Yes	Yes	Yes	No	Yes	Yes	Yes
Keras	François Chollet	MIT license	Yes	Linux, Mac OS X, Windows	Python	Python	Only if using Theano as backend	Under development for the Theano backend (and on roadmap for the TensorFlow backend)	Yes	Yes	Yes^[9]	Yes	Yes	Yes	Yes^[10]
Microsoft Cognitive Toolkit	Microsoft Research	MIT license^[11]	Yes	Windows, Linux^[12] (OSX via Docker on roadmap)	C++	Python, C++, Command line,^[13] BrainScript^[14] (.NET on roadmap^[15])	Yes^[16]	No	Yes	Yes	Yes^[17]	Yes^[18]	Yes^[18]	No^[19]	Yes^[20]
MXNet	Distributed (Deep) Machine Learning Community	Apache 2.0	Yes	Linux, Mac OS X, Windows,^[21]^[22] AWS, Android,^[23] iOS, JavaScript^[24]	Small C++ core library	C++, Python, Julia, Matlab, JavaScript, Go, R, Scala, Perl	Yes	On roadmap^[25]	Yes	Yes^[26]	Yes^[27]	Yes	Yes	Yes	Yes^[28]
Neural Designer	Artelnics	Proprietary	No	Linux, Mac OS X, Windows	C++	Graphical user interface	Yes	No	No	?	?	No	No	No	?
OpenNN	Artelnics	GNU LGPL	Yes	Cross-platform	C++	C++	Yes	No	No	?	?	No	No	No	?
TensorFlow	Google Brain team	Apache 2.0	Yes	Linux, Mac OS X, Windows^[29]	C++, Python	Python, C/C++, Java, Go	No	On roadmap^[30]^[31]	Yes	Yes^[32]	Yes^[33]	Yes	Yes	Yes	Yes
Theano	Université de Montréal	BSD license	Yes	Cross-platform	Python	Python	Yes	Under development^[34]	Yes	Yes^[35]^[36]	Through Lasagne's model zoo^[37]	Yes	Yes	Yes	Yes^[38]
Torch	Ronan Collobert, Koray Kavukcuoglu, Clement Farabet	BSD license	Yes	Linux, Mac OS X, Windows,^[39] Android,^[40] iOS	C, Lua	Lua, LuaJIT,^[41] C, utility library for C++/OpenCL^[42]	Yes	Third party implementations^[43]^[44]	Yes^[45]^[46]	Through Twitter's Autograd^[47]	Yes^[48]	Yes	Yes	Yes	Yes^[49]
Wolfram Mathematica	Wolfram Research	Proprietary	No	Windows, Mac OS X, Linux, Cloud computing	C++	Command line, Java, C++	No	Yes	Yes	Yes	Yes^[50]	Yes	Yes	Yes	Yes

^ Licenses here are a summary, and are not taken to be complete statements of the licenses. Some libraries may use other libraries internally under different licenses

Related software

Neural Engineering Object (NENGO) – A graphical and scripting software for simulating large-scale neural systems
Numenta Platform for Intelligent Computing – Numenta's open source implementation of their hierarchical temporal memory model

References

^ Atilim Gunes Baydin; Barak A. Pearlmutter; Alexey Andreyevich Radul; Jeffrey Mark Siskind (20 February 2015). "Automatic differentiation in machine learning: a survey". arXiv:1502.05767 [cs.LG].
^ "Microsoft/caffe". GitHub.
^ "OpenCL Caffe".
^ "Caffe Model Zoo".
^ "Support for Open CL · Issue #27 · deeplearning4j/nd4j". GitHub.
^ "N-Dimensional Scientific Computing for Java".
^ Chris Nicholson; Adam Gibson. "Deeplearning4j Models".
^ Deeplearning4j. "Deeplearning4j on Spark". Deeplearning4j.{{cite web}}: CS1 maint: numeric names: authors list (link)
^ https://keras.io/applications/
^ Does Keras support using multiple GPUs? · Issue #2436 · fchollet/keras
^ "CNTK/LICENSE.md at master · Microsoft/CNTK · GitHub". GitHub.
^ "Setup CNTK on your machine". GitHub.
^ "CNTK usage overview". GitHub.
^ "BrainScript Network Builder". GitHub.
^ ".NET Support · Issue #960 · Microsoft/CNTK". GitHub.
^ "How to train a model using multiple machines? · Issue #59 · Microsoft/CNTK". GitHub.
^ https://github.com/Microsoft/CNTK/issues/140#issuecomment-186466820
^ ^a ^b "CNTK - Computational Network Toolkit". Microsoft Corporation.
^ url=https://github.com/Microsoft/CNTK/issues/534
^ "Multiple GPUs and machines". Microsoft Corporation.
^ "Releases · dmlc/mxnet". Github.
^ "Installation Guide — mxnet documentation". Readthdocs.
^ "MXNet Smart Device". ReadTheDocs.
^ "MXNet.js". Github.
^ "Support for other Device Types, OpenCL AMD GPU · Issue #621 · dmlc/mxnet". GitHub.
^ http://mxnet.readthedocs.io/
^ "Model Gallery". GitHub.
^ "Run MXNet on Multiple CPU/GPUs with Data Parallel". GitHub.
^ https://developers.googleblog.com/2016/11/tensorflow-0-12-adds-support-for-windows.html
^ "tensorflow/roadmap.md at master · tensorflow/tensorflow · GitHub". GitHub.
^ "OpenCL support · Issue #22 · tensorflow/tensorflow". GitHub.
^ https://www.tensorflow.org/
^ https://github.com/tensorflow/models
^ "Using the GPU — Theano 0.8.2 documentation".
^ http://deeplearning.net/software/theano/library/gradient.html
^ https://groups.google.com/d/msg/theano-users/mln5g2IuBSU/gespG36Lf_QJ
^ "Recipes/modelzoo at master · Lasagne/Recipes · GitHub". GitHub.
^ Using multiple GPUs — Theano 0.8.2 documentation
^ https://github.com/torch/torch7/wiki/Windows
^ "GitHub - soumith/torch-android: Torch-7 for Android". GitHub.
^ "Torch7: A Matlab-like Environment for Machine Learning" (PDF).
^ "GitHub - jonathantompson/jtorch: An OpenCL Torch Utility Library". GitHub.
^ "Cheatsheet". GitHub.
^ "cltorch". GitHub.
^ "Torch CUDA backend". GitHub.
^ "Torch CUDA backend for nn". GitHub.
^ https://github.com/twitter/torch-autograd
^ "ModelZoo". GitHub.
^ https://github.com/torch/torch7/wiki/Cheatsheet#distributed-computing--parallel-processing
^ http://blog.stephenwolfram.com/2017/03/the-rd-pipeline-continues-launching-version-11-1/

[license-1] Licenses here are a summary, and are not taken to be complete statements of the licenses. Some libraries may use other libraries internally under different licenses

[2] Atilim Gunes Baydin; Barak A. Pearlmutter; Alexey Andreyevich Radul; Jeffrey Mark Siskind (20 February 2015). "Automatic differentiation in machine learning: a survey". arXiv:1502.05767 [cs.LG].

[3] "Microsoft/caffe". GitHub.

[4] "OpenCL Caffe".

[5] "Caffe Model Zoo".

[6] "Support for Open CL · Issue #27 · deeplearning4j/nd4j". GitHub.

[7] "N-Dimensional Scientific Computing for Java".

[8] Chris Nicholson; Adam Gibson. "Deeplearning4j Models".

[9] Deeplearning4j. "Deeplearning4j on Spark". Deeplearning4j.{{cite web}}: CS1 maint: numeric names: authors list (link)

[10] ttps://keras.io/applications/

[11] Does Keras support using multiple GPUs? · Issue #2436 · fchollet/keras

[12] "CNTK/LICENSE.md at master · Microsoft/CNTK · GitHub". GitHub.

[13] "Setup CNTK on your machine". GitHub.

[14] "CNTK usage overview". GitHub.

[15] "BrainScript Network Builder". GitHub.

[16] ".NET Support · Issue #960 · Microsoft/CNTK". GitHub.

[17] "How to train a model using multiple machines? · Issue #59 · Microsoft/CNTK". GitHub.

[18] ttps://github.com/Microsoft/CNTK/issues/140#issuecomment-186466820

[cntk.ai-19] "CNTK - Computational Network Toolkit". Microsoft Corporation.

[20] url=https://github.com/Microsoft/CNTK/issues/534

[21] "Multiple GPUs and machines". Microsoft Corporation.

[22] "Releases · dmlc/mxnet". Github.

[23] "Installation Guide — mxnet documentation". Readthdocs.

[24] "MXNet Smart Device". ReadTheDocs.

[25] "MXNet.js". Github.

[26] "Support for other Device Types, OpenCL AMD GPU · Issue #621 · dmlc/mxnet". GitHub.

[27] ttp://mxnet.readthedocs.io/

[28] "Model Gallery". GitHub.

[29] "Run MXNet on Multiple CPU/GPUs with Data Parallel". GitHub.

[30] ttps://developers.googleblog.com/2016/11/tensorflow-0-12-adds-support-for-windows.html

[tensorflow-roadmap-31] "tensorflow/roadmap.md at master · tensorflow/tensorflow · GitHub". GitHub.

[GitHub-32] "OpenCL support · Issue #22 · tensorflow/tensorflow". GitHub.

[33] ttps://www.tensorflow.org/

[34] ttps://github.com/tensorflow/models

[35] "Using the GPU — Theano 0.8.2 documentation".

[36] ttp://deeplearning.net/software/theano/library/gradient.html

[37] ttps://groups.google.com/d/msg/theano-users/mln5g2IuBSU/gespG36Lf_QJ

[38] "Recipes/modelzoo at master · Lasagne/Recipes · GitHub". GitHub.

[39] Using multiple GPUs — Theano 0.8.2 documentation

[40] ttps://github.com/torch/torch7/wiki/Windows

[41] "GitHub - soumith/torch-android: Torch-7 for Android". GitHub.

[42] "Torch7: A Matlab-like Environment for Machine Learning" (PDF).

[jtorch-43] "GitHub - jonathantompson/jtorch: An OpenCL Torch Utility Library". GitHub.

[44] "Cheatsheet". GitHub.

[45] "cltorch". GitHub.

[46] "Torch CUDA backend". GitHub.

[47] "Torch CUDA backend for nn". GitHub.

[48] ttps://github.com/twitter/torch-autograd

[49] "ModelZoo". GitHub.

[50] ttps://github.com/torch/torch7/wiki/Cheatsheet#distributed-computing--parallel-processing

[51] ttp://blog.stephenwolfram.com/2017/03/the-rd-pipeline-continues-launching-version-11-1/

[a]

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

Deep learning software by name

Related software

See also

References