keras-team /autokeras
AutoML library for deep learning
★ 9,341⑂ 1,385👁 288⑃ 909● 919▤ 60
+0
Daily
9,341
stars
1,385
forks
288
watchers
909
prs
919
issues
60
releases
README
Official Website: autokeras.com
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 to everyone.
Learning resources
A short example.
python
import autokeras as ak
clf = ak.ImageClassifier()
clf.fit(x_train, y_train)
results = clf.predict(x_test)
Official website tutorials.
The book of Automated Machine Learning in Action.
The LiveProjects of Image Classification with AutoKeras.
 
 
Installation
To install the package, please use the pip installation as follows:
shell
pip3 install autokeras
Please follow the installation guide for more details.
Note: Currently, AutoKeras is only compatible with Python = 3.7 and TensorFlow = 2.8.0.
Community
Ask your questions on our GitHub Discussions.
Contributing Code
Here is how we manage our project.
We pick the critical issues to work on from GitHub issues.
They will be added to this Project.
Some of the issues will then be added to the milestones,
which are used to plan for the releases.
Refer to our Contributing Guide to learn the best practices.
Thank all the contributors!
Cite this work
Haifeng Jin, François Chollet, Qingquan Song, and Xia Hu. "AutoKeras: An AutoML Library for Deep Learning." the Journal of machine Learning research 6 (2023): 1-6. (Download)
Biblatex entry:
bibtex
@article{JMLR:v24:20-1355,
author = {Haifeng Jin and François Chollet and Qingquan Song and Xia Hu},
title = {AutoKeras: An AutoML Library for Deep Learning},
journal = {Journal of Machine Learning Research},
year = {2023},
volume = {24},
number = {6},
pages = {1--6},
url = {http://jmlr.org/papers/v24/20-1355.html}
}
Acknowledgements
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.Top issues
#1420Specify hyperparameter ranges for blocks20 comments · 2020-11-05#1513AutoKeras + Deep Metrics Learning (i.e. triplet loss)15 comments · 2021-02-20#1915Bug: StructuredDataClassifier and StructuredDataRegressor gone?12 comments · 2024-04-10#1091ValueError: Input contains NaN, infinity or a value too large for dtype('float64')12 comments · 2020-04-09#925Object Detection12 comments · 2020-01-24#1298Autokeras slow to start for large dataset?11 comments · 2020-08-21#1906Bug: import autokeras as ak --> ModuleNotFoundError: No module named 'tensorflow.keras.layers.experimental'10 comments · 2024-03-12#1479max_model_size is not useful with ImageClassifier10 comments · 2021-01-04
Discussions
#1948Extending data for training0 comments · 2025-08-10#1946ValueError: The name "resnet50" is used 4 times in the model. All operation names should be unique.0 comments · 2025-04-23#1937Is Adam the end of Gradient descent (and model Optimizers)?0 comments · 2024-11-12#1936Improving Activation Function0 comments · 2024-11-12#1935Allowing the setting of sizes for `DenseBlock` (and possibly other blocks)0 comments · 2024-11-12#1934Adding Granularity to Dropout Values0 comments · 2024-11-12#1933Alternative Tools like Deep_AutoViML0 comments · 2024-11-07#1932Not getting the new installs working0 comments · 2024-11-07
Latest release
Release v3.0.0 3.0.0 2025-11-12This new release repositioned AutoKeras as a convenient library for quick experiments on small datasets with simple deep learning models, rather than a large-scale training library for distributed settings. What's Changed AutoKeras works on PyTorch, JAX, and TensorFlow. PyTorch is the recommended backend. Added back StructuredDataClassifier, StructuredDataRegressor, StructuredInput, StructuredDataBlock, Embedding. Website updated with new documentation. Breaking Changes Removed BertBlock. Removed support for using tensorflow.data.Dataset, pandas.DataFrame, strings for file pa