Home

Awesome

<p align="center"> <img src="https://raw.githubusercontent.com/DataCanvasIO/Hypernets/master/docs/source/images/Hypernets.png" width="500" >

Python Versions Downloads PyPI Version

We Are Hiring!

Dear folks, we are offering challenging opportunities located in Beijing for both professionals and students who are keen on AutoML/NAS. Come be a part of DataCanvas! Please send your CV to yangjian@zetyun.com. (Application deadline: TBD.)

Hypernets: A General Automated Machine Learning Framework

Hypernets is a general AutoML framework, based on which it can implement automatic optimization tools for various machine learning frameworks and libraries, including deep learning frameworks such as tensorflow, keras, pytorch, and machine learning libraries like sklearn, lightgbm, xgboost, etc. It also adopted various state-of-the-art optimization algorithms, including but not limited to evolution algorithm, monte carlo tree search for single objective optimization and multi-objective optimization algorithms such as MOEA/D,NSGA-II,R-NSGA-II. We introduced an abstract search space representation, taking into account the requirements of hyperparameter optimization and neural architecture search(NAS), making Hypernets a general framework that can adapt to various automated machine learning needs. As an abstraction computing layer, tabular toolbox, has successfully implemented in various tabular data types: pandas, dask, cudf, etc.

Overview

Conceptual Model

<p align="center"> <img src="https://raw.githubusercontent.com/DataCanvasIO/Hypernets/master/docs/source/images/hypernets_conceptual_model.png" width="100%"/> </p>

Illustration of the Search Space

<p align="center"> <img src="https://raw.githubusercontent.com/DataCanvasIO/Hypernets/master/docs/source/images/hypernets_search_space.png" width="100%"/> </p>

What's NEW !

Installation

Conda

Install Hypernets with conda from the channel conda-forge:

conda install -c conda-forge hypernets

Pip

Install Hypernets with different options:

pip install hypernets
pip install hypernets[notebook]
pip install hypernets[dask]
pip install hypernets[zhcn]
pip install hypernets[all]

To Verify your installation:

python -m hypernets.examples.smoke_testing

Related Links

Documents

Neural Architecture Search

Hypernets related projects

DataCanvas AutoML Toolkit

Citation

If you use Hypernets in your research, please cite us as follows:

Jian Yang, Xuefeng Li, Haifeng Wu. Hypernets: A General Automated Machine Learning Framework. https://github.com/DataCanvasIO/Hypernets. 2020. Version 0.2.x.

BibTex:

@misc{hypernets,
  author={Jian Yang, Xuefeng Li, Haifeng Wu},
  title={{Hypernets}: { A General Automated Machine Learning Framework}},
  howpublished={https://github.com/DataCanvasIO/Hypernets},
  note={Version 0.2.x},
  year={2020}
}

DataCanvas

Hypernets is an open source project created by DataCanvas.