VSG³A² : A Genetic Algorithm-Based Virtual Sample Generation Approach Using Information Gain and Acceptance-Rejection Sampling

Paper title: VSG³A² : A Genetic Algorithm-Based Virtual Sample Generation Approach Using Information Gain and Acceptance-Rejection Sampling

by Hong Yu and Xuekang Fan et al.

Copyright (C) 2023 Hong Yu Chongqing University of Posts and Telecommunications

Usage

• Python 3.9
• Numpy, pytorch, sklearn, random, math, matplotlib

Describe

The repositories include five folders. Each folder corresponds to one of the experiments in this paper.

1. The folder “train_model” provides the source codes of our VSG model, in which the the "my_Model.py" file is the starter program and you will need to modify the internal code to achieve specific application effects.
2. The folder “data” provides the source codes of our test model, specifically, this is a single hidden layer back propagation neural network.
3. The folder "data" contains a number of data files, including experimental data and program logs.

Experimental environment:

Pycharm 2022.1.2 Edition, Intel(R) Core(TM) i5-8500 CPU @ 3.00 GHz, 16.0 GB RAM

Virtual environment:

python 3.9, numpy 1.21.5, pytorch 1.11.0, sklearn 1.1.1, matplotlib 3.5.1

reference

Hong Yu, Xuekang Fan, Guoyin Wang, Yongfang Xie. VSG³A² : A Genetic Algorithm-Based Virtual Sample Generation Approach Using Information Gain and Acceptance-Rejection Sampling.

This work has been submitted to IEEE Transactions on Evolutionary Computation