/squamish_experiments

Primary LanguagePython

Squamish Experiments

This repository contains experiments for the scientific publication about Squamish. Squamish is a new feature selection tool to find all-relevant features and their relevance classes.

@misc{pfannschmidt2020sequential,
    title={Sequential Feature Classification in the Context of Redundancies},
    author={Lukas Pfannschmidt and Barbara Hammer},
    year={2020},
    eprint={2004.00658},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

Preprints can be found at https://pub.uni-bielefeld.de/record/2942271 or https://arxiv.org/abs/2004.00658.

Replicate Experiments

To replicate the experimental results of the paper (figure and tables) we provide a docker image and several scripts to produce a (hopefully) identical output.

1. Get Image

Build the image yourself with

docker build -t squamish_experiments .

or pull it from DockerHub

docker pull mirek1337/squamish_experiments

2. Run container

docker run -v ./tmp:/exp/tmp:Z -v ./output:/exp/output:Z -it squamish_experiments make

which calls make inside the container to execute all experiments in the Makefile.

After the experiments are done (can take several hours) the output should end up in the ./output folder.

It's possible to change the following parameters as environment variables in the docker command via the -e option:

Defaults used in paper

  • SEED = 123
  • REPEATS = 10
  • N_THREADS = 1