https://dx.doi.org/10.1002/cem.695

[vfonov]

I implemented method Johan Trygg and Svante Wold. Orthogonal projections to latent structures (O-PLS). J. Chemometrics 2002; 16: 119-128. DOI: [10.1002/cem.695]

It's in the form of pre-filtering step that can be applied before executing pls:

model_opls = PartialLeastSquaresRegressor.OPLS1Model(X,Y,n_ortho)
X_flt, X_ortho=PartialLeastSquaresRegressor.fitting(model_opls,X,Y)
...
model_pls = PartialLeastSquaresRegressor.PLSModel(X_flt,Y,n_fact)
PartialLeastSquaresRegressor.fitting(model_pls, X_flt, Y)
...