This package implements a method for minimum sample size estimation in the training step of several supervised machine learning algorithms for classification: kNN, logistic regression, naive Bayes, and random forest.
The main function is called "MinSizeClassification".
All documentation is included in the function description.
On an R console or RStudio, execute the following line:
devtools::install_github("gpcastelo/MinSizeML")
This is the repository for the master thesis at the Master's degree of Bioinformatics and Biostatistics, Universitat Oberta de Catalunya.
Authored by Guillermo Prol Castelo and supervised by Jose Luis Mosquera Mayo.