A lightweight and visual AutoML system
Cooka is a lightweight and visualization toolkit to manage datasets and design model learning experiments through web UI. It's using DeepTables and HyperGBM as experiment engine to complete feature engineering, neural architecture search and hyperparameter tuning automatically.
Through the web UI provided by cooka you can:
Screen shots:
The machine learning algorithms supported are :
The neural networks supported are:
The search algorithms supported are:
The supported feature engineering provided by scikit-learn and featuretools are:
Scaler
Encoder
Discretizer
Dimension Reduction
Feature derivation
Missing value filling
It can also extend the search space to support more feature engineering methods and modeling algorithms.
The python version should be >= 3.6, for CentOS , install the system package:
pip install --upgrade pip
pip install cooka
Start the web server:
cooka server
Then open http://<your_ip:8000>
with your browser to use cooka.
By default, the cooka configuration file is at ~/.config/cooka/cooka.py
, to generate a template:
mkdir -p ~/.config/cooka/
cooka generate-config > ~/.config/cooka/cooka.py
Launch a Cooka docker container:
docker run -ti -p 8888:8888 -p 8000:8000 -p 9001:9001 -e COOKA_NOTEBOOK_PORTAL=http://<your_ip>:8888 datacanvas/cooka:latest
Open http://<your_ip:8000>
with your browser to visit cooka.
If you use Cooka in your research, please cite us as follows:
Haifeng Wu, Jian Yang. Cooka: A lightweight and visual AutoML system. https://github.com/DataCanvasIO/Cooka, 2021. Version 0.1.x
@misc{cooka,
author={Haifeng Wu, Jian Yang},
title={{Cooka}: {A lightweight and visual AutoML system}},
howpublished={https://github.com/DataCanvasIO/Cooka},
note={Version 0.1.x},
year={2021}
}
Cooka is an open source project created by DataCanvas.