Learn R Programming

TPLSr 1.0.4

R Package for using Thresholded Partial Least Squares (TPLS) for big data regression and classification. It is developed with whole-brain neuroimaging (fMRI) MVPA predictors in mind. TPLS uses analytical calulations of partial least squares to dramatically speed-up the training of models with large number of features (~millions).

Citation: Lee, S., Bradlow, E. T., & Kable, J. W. (2022). Fast construction of interpretable whole-brain decoders. Cell Reports Methods. doi: https://doi.org/10.1016/j.crmeth.2022.100227

CRAN: https://CRAN.R-project.org/package=TPLSr
GITHUB: https://github.com/sangillee/TPLSr

Copy Link

Version

Install

install.packages('TPLSr')

Monthly Downloads

174

Version

1.0.4

License

GPL-3

Maintainer

Sangil Lee

Last Published

June 10th, 2022

Functions in TPLSr (1.0.4)

makePredictor

Method for extracting the T-PLS predictor at a given compval and threshval
TPLSdat

Sample participant data from a left-right button press task
TPLSpredict

Method for making predictions on a testing dataset testX
plotTuningSurface

Plots the tuning surface of TPLS
TPLS_cv

Constructor method for fitting a cross-validation T-PLS model
TPLS

Constructor method for fitting a T-PLS model with given data X and Y.
evalTuningParam

Evaluating cross-validation performance of a TPLS_cv model at compvec and threshvec