Metadata-Version: 2.4
Name: auto-opt
Version: 0.0.2
Summary: An AutoDL Package
Author-email: Nick Cliffel <cliffel.11@osu.edu>
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: tapipy>=1.0.0
Requires-Dist: joblib>=1.3.0
Requires-Dist: matplotlib>=3.5.0
Requires-Dist: netext>=0.5.0
Requires-Dist: networkx>=3.0
Requires-Dist: numpy<3.0.0,>=1.22.0
Requires-Dist: pandas>=2.0.0
Requires-Dist: psutil>=5.9.0
Requires-Dist: pyro_ppl>=1.8.0
Requires-Dist: rich>=13.0.0
Requires-Dist: scikit-learn>=1.2.0
Requires-Dist: threadpoolctl>=3.1.0
Requires-Dist: torch>=2.0.0
Requires-Dist: torch_geometric>=2.3.0
Requires-Dist: tqdm>=4.64.0
Dynamic: license-file

# auto-opt-package

## Name

auto-opt

## Description

This is an opensource AutoDL framework for tabular datasets. auto-opt will provide access a variety of different optimization methods in the future, however, currently auto-opt is centered around hill climbing with multiple step sizes. This form of optimization while simple produces good results while easily handeling categorical values.

## Visuals

Screenshots will be added at a later data.

## Installation

This code based can be downloaded via gitlab at (https://code.osu.edu/cliffel.11/auto-opt-package) or can be installed and utilized via pip.

## Usage

Usage examples will be added at a later date.

## Support

For bug reports and other issues please contact cliffel.11@osu.edu

## Roadmap

In the coming months we will finish the BayesOpt optimizer and add more node types. While the GNN based search is laid out it has not been throughly tested but will be in the future.

## Contributing

Right now we are not accepting contributions from others due to contribution requirements not having been established but hope to in the future.

## Authors and acknowledgment

Nick Cliffel (cliffel.11@osu.edu)
Rajiv Ramnath

National Science Foundation (NSF) funded AI institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment (ICICLE) (OAC 2112606)

## License

This project is licensed under the MIT License. See the `LICENSE.txt` file for details.

## Project status

The project is currently undergoing development. The hill climbing based optimizers are stable and can be used freely. The BayesOpt optimizer is still incomplete at this time, however, is included in the release as it shows how to use the developed models for inference. The GNN search should has not been throughly tested and is still relatively unstable.
