preface

Medium writer Mybridge has picked 10 GitHub machine learning projects out of 8,800 that have received more than a few stars

No 1 FastText

Library for fast text representation and classification.

Github: 16,510 stars.

Link: https://goo.gl/VWGfCs

Note: Muse, a multilingual unsupervised or supervised word embedding project based on FastText, has 695 stars on GitHub

No 2 Deep-photo-styletransfer

The code and data of “deep photo style conversion”.

Github: 8,484 stars.

Link: https://github.com/luanfujun/deep-photo-styletransfer

No 3 The world’s simplest face recognition API, based on Python with command line.

Github: 18,875 stars.

Link: https://github.com/ageitgey/face_recognition

No 4 Magenta

Music and art generated by machine intelligence

Github: 11626 stars.

Link: https://github.com/tensorflow/magenta

No 5 Sonnet

Neural network library based on TensorFlow

Github: 7,049 stars.

Link: https://github.com/deepmind/sonnet

No 6 TensorFlow.js

Hardware-accelerated machine intelligence library for networking

Github: 8,163 stars.

Link: https://github.com/PAIR-code/deeplearnjs

No 7

Fast style transfer based on TensorFlow

GitHub: 7,028 stars

Link: https://github.com/lengstrom/fast-style-transfer

No 8 Pysc2

Starcraft II learning environment

Github: 5,220 stars.

Link: https://github.com/deepmind/pysc2

No 9 AirSim

Open source simulator based on Unreal Engine for Microsoft AI and Research’s autonomous vehicles

Github: 6,334 stars

From Shital Shah, Microsoft

Link: https://github.com/Microsoft/AirSim?

No 10 Facets

Visualization of machine learning data sets

Github: 4,969 stars

Link: https://github.com/PAIR-c