⚠️ NOTE: This page describes a new model assessment and tuning workflow in the development version of mixOmics. We invite you to try it and share feedback during beta testing! To install, use:devtools::install_github("mixOmicsTeam/mixOmics", ref = "6.31.4")
If you’re using the stable Bioconductor version, this page won’t apply.
This page provides guidance on using the tune() function in mixOmics for parameter tuning, essential for optimising model performance. Parameter tuning helps with optimal selection of number of components (ncomp), distance metrics (dist), and variables to retain (keepX/keepY). The tune() function uses cross-validation to identify the best-performing parameter combinations. This page offers examples for tuning regression models (PLS, sPLS) and classification models (PLS-DA, sPLS-DA), and special cases like PCA and CCA. This process enables effective parameter tuning while balancing computational efficiency.
Data used on this page:liver.toxicitybreast.tumorsnutrimouse
Key functions used on this page:tune()tune.pls()tune.spls()tune.plsda()tune.splsda()tune.pca()tune.rcc()