aldantas / MLB_AS_QAP_BRACIS2018

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Experiment Materials

This repository stores the experiments materials from the paper "Selecting Algorithms for the Quadratic Assignment Problem with a Multi-label Meta-learning Approach" to be pusblished at the Proceedings of the The 7th Brazilian Conference on Intelligent Systems (BRACIS).

The content of each folder is described next.

Algorithms

This folder contains the source-codes of the meta-heuristics with their default parameters. It is also provided the set of seeds used for the experiments and the runner scripts for the landscape features extraction. Besides, the fitted function used to determine the execution time based on the instance size can be found here.

Algorithms Performances

The results achieved by the algorithms over 30 runs on each intance.

Dataset

The generated multi-label classification dataset. The labels are integer encoded and must be binarized before using them with the scikit-lean methods.

Reults

It contains the results metrics, such as the Accuracy and F-Scores, of both techniques performed for handling the multi-label classification.

QAP Intances

These are all the instances retrieved from QAPLIB, along with their respective best known solutions.

About


Languages

Language:C 79.2%Language:C++ 13.1%Language:Python 6.8%Language:Makefile 0.8%Language:Shell 0.0%