Description: libPLS is an integrated library for facilitating the use of PLS written in MATLAB.  It includes a whole set of functions for data pretreatment, outlier detection, variable selection, model building, cross validation and prediction.


CitationLi H.-D., Xu Q.-S., Liang Y.-Z. (2014) libPLS: An Integrated Library for Partial Least Squares Regression and Discriminant Analysis. PeerJ PrePrints 2:e190v1, source codes available at



8 thoughts on “libPLS

  1. parisa

    It has been a while since I have started surfing the net for finding a matlab code to perform Opls! once I came across to this toolbox, it was really unbelievable for me. Softwares like Unscrambler and Simca were not as advantageous as this toolbox was for me!
    Showing my appreciation is the least thing I could do.
    Thank you a million 100000000000000000000000000000000 😉

    1. Alin

      could you tell me the performance of the libpls toolbox compared with that available in off the shelf packages such as The Unscrambler, please?

  2. Kishore

    How can i save the created model from the PLS? I want to use the saved model for unknown sample. Is there any way to save the model changing th following lines:

    PLS=pls(Xcal,ycal,10); %+++ Build a PLS regression model using training set
    [ypred,RMSEP]=plsval(PLS,Xtest,ytest); %+++ make predictions on test set

    Looking forward for your reply,

    Thanks & Regards,


  3. qianqian

    I would like to ask how libplsda for multi-classification, multi-categorical variables how to filter, thank you.


Leave a Reply

Your email address will not be published. Required fields are marked *