Skip to content


Let's Chat

To make AutoKeras better, I would like to hear your thoughts. I am happy to answer any questions you have about our project. Join our Slack and send me (Haifeng Jin) a message. I will schedule a meeting with you.

AutoKeras: An AutoML system based on Keras. It is developed by DATA Lab at Texas A&M University. The goal of AutoKeras is to make machine learning accessible for everyone.


Here is a short example of using the package.

import autokeras as ak

clf = ak.ImageClassifier(), y_train)
results = clf.predict(x_test)

For detailed tutorial, please check here.


To install the package, please use the pip installation as follows:

pip3 install git+
pip3 install autokeras==1.0.5

Please follow the installation guide for more details.

Note: Currently, AutoKeras is only compatible with Python >= 3.5 and TensorFlow >= 2.3.0.


Slack: Request an invitation. Use the #autokeras channel for communication.

Twitter: You can also follow us on Twitter @autokeras for the latest news.

Emails: Subscribe our email list to receive announcements.

Online Meetings: Join the Google group and our online meetings will appear on your Google Calendar.

QQ Group: Join our QQ group 1150366085. Password: akqqgroup


You can follow the Contributing Guide to become a contributor.

If you don't know where to start, please join our community on Slack and ask us. We will help you get started!

Thank all the contributors!


We accept financial support on Open Collective. Thank every backer for supporting us!

Cite this work

Haifeng Jin, Qingquan Song, and Xia Hu. "Auto-keras: An efficient neural architecture search system." Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. ACM, 2019. (Download)

Biblatex entry:

  title={Auto-Keras: An Efficient Neural Architecture Search System},
  author={Jin, Haifeng and Song, Qingquan and Hu, Xia},
  booktitle={Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery \& Data Mining},


The authors gratefully acknowledge the D3M program of the Defense Advanced Research Projects Agency (DARPA) administered through AFRL contract FA8750-17-2-0116; the Texas A&M College of Engineering, and Texas A&M University.