Within or Cross Group Cross-Validation

Owned by Star Yanxin Gao
May 12, 2020
2 min read


Created by Juan David Arbelaez Velez , last modified by Star Yanxin Gao on Aug 31, 2018

Step 01. Get the data. 

Three different files in .tab format are required to be uploaded, a genotype, phenotype and group data file. For reference please see the example files "genoForBGLRf.tab", "phenotypesForBGLRTestV2.tab", and "grpsBGLRTest.tab" respectively. 

For the phenotype file you must select the column that contains the samples IDs, and the column with the phenotype to be analyzed.



Step 02. Choose the type of cross-validation. 

There are currently two options supported, within and across groups.



Step 03. Choose the k value and models to be analyzed. 

Write the number of folds to perform, and select the different models to be tested, currently Bayes A, Bayes B, Bayes C, Bayesian Lasso, and BRR.







References

  1. Brownlee, Jason. A Gentle Introduction to k-fold Cross-Validation (2018). Source: https://machinelearningmastery.com/k-fold-cross-validation/