Skip to content

Source code for the Paper: CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints}

License

Notifications You must be signed in to change notification settings

martius-lab/CombOptNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints

Architecture overview

This repository contains PyTorch implementation of the paper CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints

Installation

  1. Run pipenv install (at your own risk with --skip-lock to save some time).
  2. From within the pipenv environment run python3 -m pip install -i https://pypi.gurobi.com gurobipy.
  3. Obtain a license and download/set it.
  4. Download and extract the datasets.

Usage

For [experiment] = knapsack or [experiment] = static_constraints:

  1. Set the base_dataset_path parameter in experiments/[experiment]/base.yaml.
  2. In case of static constraints: set the dataset_specification parameter in experiments/static_constraints/base.yaml
  3. Run python3 main.py experiments/[experiment]/[method].yaml.

Citation

@misc{paulus2021comboptnet,
      title={CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints}, 
      author={Anselm Paulus and Michal Rolínek and Vít Musil and Brandon Amos and Georg Martius},
      year={2021},
      eprint={2105.02343},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

About

Source code for the Paper: CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints}

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages