Comparison of deep learning software

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

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


Deep learning software by name[edit]

Software Creator Software license[a] Open source Platform Written in Interface OpenMP support OpenCL support CUDA support Parallel execution (multi node) Automatic differentiation[1] Has pretrained models Recurrent nets Convolutional nets RBM/DBNs Metal support
roNNie.ai Kevin Lok MIT Yes Linux, macOS, Windows Python Python Yes Yes Yes Yes
BigDL Jason Dai Apache 2.0 Yes Apache Spark Scala Scala, Python No Yes Yes Yes
Caffe Berkeley Vision and Learning Center BSD license Yes Linux, macOS, Windows[2] C++ Python, MATLAB, C++ Yes Under development[3] Yes Yes Yes[4] Yes Yes Yes ?
Deeplearning4j Skymind engineering team; Deeplearning4j community; originally Adam Gibson Apache 2.0 Yes Linux, macOS, Windows, Android (Cross-platform) C++, Java Java, Scala, Clojure, Python (Keras), Kotlin Yes On roadmap[5] Yes[6][7] Computational Graph Yes[8] Yes Yes Yes Yes[9]
Chainer Preferred Networks MIT license Yes Linux, macOS, Windows Python No No[10][11] Yes Yes Yes Yes Yes
Darknet Joseph Redmon Public Domain Yes Cross-Platform C C, Python Yes No[12] Yes Yes
Dlib Davis King Boost Software License Yes Cross-Platform C++ C++ Yes No Yes Yes Yes Yes Yes Yes Yes
DataMelt (DMelt) S.Chekanov Freemium Yes Cross-Platform Java Java No No No No No No No No No
DyNet Carnegie Mellon University Apache 2.0 Yes Linux, macOS, Windows C++, Python No[13] Yes Yes Yes
Intel Data Analytics Acceleration Library Intel Apache License 2.0 Yes Linux, macOS, Windows on Intel CPU[14] C++, Python, Java C++, Python, Java[14] Yes No No Yes No Yes Yes
Intel Math Kernel Library Intel Proprietary No Linux, macOS, Windows on Intel CPU[15] C[16] Yes[17] No No Yes No Yes[18] Yes[18] No
Keras François Chollet MIT license Yes Linux, macOS, Windows Python Python, R Only if using Theano as backend Under development for the Theano backend (and on roadmap for the TensorFlow backend) Yes Yes Yes[19] Yes Yes Yes Yes[20]
MATLAB + Neural Network Toolbox MathWorks Proprietary No Linux, macOS, Windows C, C++, Java, MATLAB MATLAB No No Train with Parallel Computing Toolbox and generate CUDA code with GPU Coder[21] No Yes[22][23] Yes[22] Yes[22] Yes[22] With Parallel Computing Toolbox[24]
Microsoft Cognitive Toolkit Microsoft Research MIT license[25] Yes Windows, Linux[26] (macOS via Docker on roadmap) C++ Python (Keras), C++, Command line,[27] BrainScript[28] (.NET on roadmap[29]) Yes[30] No Yes Yes Yes[31] Yes[32] Yes[32] Yes Yes[33]
Apache MXNet Apache Software Foundation Apache 2.0 Yes Linux, macOS, Windows,[34][35] AWS, Android,[36] iOS, JavaScript[37] Small C++ core library C++, Python, Julia, Matlab, JavaScript, Go, R, Scala, Perl Yes On roadmap[38] Yes Yes[39] Yes[40] Yes Yes Yes Yes[41]
Neural Designer Artelnics Proprietary No Linux, macOS, Windows C++ Graphical user interface Yes No No ? ? No No No ?
OpenNN Artelnics GNU LGPL Yes Cross-platform C++ C++ Yes No Yes ? ? No No No ?
PlaidML Vertex.AI AGPL3 Yes Linux, macOS, Windows C++, Python Keras, Python, C++, C No Yes Yes Yes Yes Yes Yes ? Yes
PyTorch Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan BSD license Yes Linux, macOS, Windows Python, C, CUDA Python Yes Via separately maintained package[42][43][44] Yes Yes Yes Yes Yes Yes
Apache SINGA Apache Incubator Apache 2.0 Yes Linux, macOS, Windows C++ Python, C++, Java No No Yes ? Yes Yes Yes Yes Yes
TensorFlow Google Brain team Apache 2.0 Yes Linux, macOS, Windows,[45] Android C++, Python, CUDA Python (Keras), C/C++, Java, Go, R[46], Julia No On roadmap[47] but already with SYCL[48] support Yes Yes[49] Yes[50] Yes Yes Yes Yes
TensorLayer Hao Dong Apache 2.0 Yes Linux, macOS, Windows,[51] Android C++, Python, Python No On roadmap[47] but already with SYCL[48] support Yes Yes[52] Yes[53] Yes Yes Yes Yes
Theano Université de Montréal BSD license Yes Cross-platform Python Python (Keras) Yes Under development[54] Yes Yes[55][56] Through Lasagne's model zoo[57] Yes Yes Yes Yes[58]
Torch Ronan Collobert, Koray Kavukcuoglu, Clement Farabet BSD license Yes Linux, macOS, Windows,[59] Android,[60] iOS C, Lua Lua, LuaJIT,[61] C, utility library for C++/OpenCL[62] Yes Third party implementations[63][64] Yes[65][66] Through Twitter's Autograd[67] Yes[68] Yes Yes Yes Yes[69]
Wolfram Mathematica Wolfram Research Proprietary No Windows, macOS, Linux, Cloud computing C++, Wolfram Language, CUDA Wolfram Language Yes No Yes Yes Yes[70] Yes Yes Yes Under Development
VerAI VerAI Proprietary No Linux, Web-based C++,Python, Go, Angular Graphical user interface, cli No No Yes Yes Yes Yes Yes Yes Yes
  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

Related software[edit]

See also[edit]

References[edit]

  1. ^ Atilim Gunes Baydin; Barak A. Pearlmutter; Alexey Andreyevich Radul; Jeffrey Mark Siskind (20 February 2015). "Automatic differentiation in machine learning: a survey". arXiv:1502.05767Freely accessible [cs.LG]. 
  2. ^ "Microsoft/caffe". GitHub. 
  3. ^ "OpenCL Caffe". 
  4. ^ "Caffe Model Zoo". 
  5. ^ "Support for Open CL · Issue #27 · deeplearning4j/nd4j". GitHub. 
  6. ^ "N-Dimensional Scientific Computing for Java". 
  7. ^ "Comparing Top Deep Learning Frameworks". Deeplearning4j. 
  8. ^ Chris Nicholson; Adam Gibson. "Deeplearning4j Models". 
  9. ^ Deeplearning4j. "Deeplearning4j on Spark". Deeplearning4j. 
  10. ^ https://github.com/chainer/chainer/pull/2717
  11. ^ https://github.com/chainer/chainer/issues/99
  12. ^ https://github.com/pjreddie/darknet/issues/127
  13. ^ https://github.com/clab/dynet/issues/405
  14. ^ a b Intel® Data Analytics Acceleration Library (Intel® DAAL) | Intel® Software
  15. ^ Intel® Math Kernel Library (Intel® MKL) | Intel® Software
  16. ^ Deep Neural Network Functions
  17. ^ Using Intel® MKL with Threaded Applications | Intel® Software
  18. ^ a b Intel® Xeon Phi™ Delivers Competitive Performance For Deep Learning—And Getting Better Fast | Intel® Software
  19. ^ https://keras.io/applications/
  20. ^ Does Keras support using multiple GPUs? · Issue #2436 · fchollet/keras
  21. ^ "GPU Coder - MATLAB & Simulink". MathWorks. Retrieved 13 November 2017. 
  22. ^ a b c d "Neural Network Toolbox - MATLAB". MathWorks. Retrieved 13 November 2017. 
  23. ^ "Deep Learning Models - MATLAB & Simulink". MathWorks. Retrieved 13 November 2017. 
  24. ^ "Parallel Computing Toolbox - MATLAB". MathWorks. Retrieved 13 November 2017. 
  25. ^ "CNTK/LICENSE.md at master · Microsoft/CNTK · GitHub". GitHub. 
  26. ^ "Setup CNTK on your machine". GitHub. 
  27. ^ "CNTK usage overview". GitHub. 
  28. ^ "BrainScript Network Builder". GitHub. 
  29. ^ ".NET Support · Issue #960 · Microsoft/CNTK". GitHub. 
  30. ^ "How to train a model using multiple machines? · Issue #59 · Microsoft/CNTK". GitHub. 
  31. ^ https://github.com/Microsoft/CNTK/issues/140#issuecomment-186466820
  32. ^ a b "CNTK - Computational Network Toolkit". Microsoft Corporation. 
  33. ^ "Multiple GPUs and machines". Microsoft Corporation. 
  34. ^ "Releases · dmlc/mxnet". Github. 
  35. ^ "Installation Guide — mxnet documentation". Readthdocs. 
  36. ^ "MXNet Smart Device". ReadTheDocs. 
  37. ^ "MXNet.js". Github. 
  38. ^ "Support for other Device Types, OpenCL AMD GPU · Issue #621 · dmlc/mxnet". GitHub. 
  39. ^ https://mxnet.readthedocs.io/
  40. ^ "Model Gallery". GitHub. 
  41. ^ "Run MXNet on Multiple CPU/GPUs with Data Parallel". GitHub. 
  42. ^ https://github.com/hughperkins/pytorch-coriander
  43. ^ https://github.com/pytorch/pytorch/issues/488
  44. ^ https://github.com/pytorch/pytorch/issues/488#issuecomment-273626736
  45. ^ https://developers.googleblog.com/2016/11/tensorflow-0-12-adds-support-for-windows.html
  46. ^ interface), JJ Allaire (R; RStudio; Eddelbuettel, Dirk; Golding, Nick; Tang, Yuan; Tutorials), Google Inc (Examples and (2017-05-26), tensorflow: R Interface to TensorFlow, retrieved 2017-06-14 
  47. ^ a b "tensorflow/roadmap.md at master · tensorflow/tensorflow · GitHub". GitHub. January 23, 2017. Retrieved May 21, 2017. 
  48. ^ a b "OpenCL support · Issue #22 · tensorflow/tensorflow". GitHub. 
  49. ^ https://www.tensorflow.org/
  50. ^ https://github.com/tensorflow/models
  51. ^ https://developers.googleblog.com/2016/11/tensorflow-0-12-adds-support-for-windows.html
  52. ^ https://www.tensorflow.org/
  53. ^ https://github.com/tensorflow/models
  54. ^ "Using the GPU — Theano 0.8.2 documentation". 
  55. ^ http://deeplearning.net/software/theano/library/gradient.html
  56. ^ https://groups.google.com/d/msg/theano-users/mln5g2IuBSU/gespG36Lf_QJ
  57. ^ "Recipes/modelzoo at master · Lasagne/Recipes · GitHub". GitHub. 
  58. ^ Using multiple GPUs — Theano 0.8.2 documentation
  59. ^ https://github.com/torch/torch7/wiki/Windows
  60. ^ "GitHub - soumith/torch-android: Torch-7 for Android". GitHub. 
  61. ^ "Torch7: A Matlab-like Environment for Machine Learning" (PDF). 
  62. ^ "GitHub - jonathantompson/jtorch: An OpenCL Torch Utility Library". GitHub. 
  63. ^ "Cheatsheet". GitHub. 
  64. ^ "cltorch". GitHub. 
  65. ^ "Torch CUDA backend". GitHub. 
  66. ^ "Torch CUDA backend for nn". GitHub. 
  67. ^ https://github.com/twitter/torch-autograd
  68. ^ "ModelZoo". GitHub. 
  69. ^ https://github.com/torch/torch7/wiki/Cheatsheet#distributed-computing--parallel-processing
  70. ^ http://resources.wolframcloud.com/NeuralNetRepository