Regression and Similarity Evaluation for Memory-Based Learning in Spectral Chemometrics

Leo Ramirez-Lopez & Antoine Stevens

Visit the resemble site here

Installing the package is very simple:

install.packages('resemble')

If you do not have the following packages installed, in some cases it is better to install them first

install.packages('Rcpp')
install.packages('RcppArmadillo')
install.packages('foreach')
install.packages('iterators')

Note: Apart from these packages we stronly recommend to download and install Rtools (directly from here or from CRAN https://cran.r-project.org/bin/windows/Rtools/). This is important for obtaining the proper C++ toolchain that you might need for using resemble.

Then, install resemble

install.packages('C:/MyFolder/resemble-1.2.2.zip', repos = NULL)

The development version can be obtained at the package website

After installing resemble you should be also able to run the following lines:

require(resemble)

help(mbl)

#install.packages('prospectr')
require(prospectr)

data(NIRsoil)

Xu <- NIRsoil$spc[!as.logical(NIRsoil$train),]
Yu <- NIRsoil$CEC[!as.logical(NIRsoil$train)]
Yr <- NIRsoil$CEC[as.logical(NIRsoil$train)]
Xr <- NIRsoil$spc[as.logical(NIRsoil$train),]

Xu <- Xu[!is.na(Yu),]
Xr <- Xr[!is.na(Yr),]

Yu <- Yu[!is.na(Yu)]
Yr <- Yr[!is.na(Yr)]

# Example of the mbl function
# A mbl approach (the spectrum-based learner) as implemented in Ramirez-Lopez et al. (2013)
# An exmaple where Yu is supposed to be unknown, but the Xu (spectral variables) are known
ctrl <- mblControl(sm = 'pc', pcSelection = list('opc', 40),
                   valMethod = 'NNv', center = TRUE)

sbl.u <- mbl(Yr = Yr, Xr = Xr, Yu = NULL, Xu = Xu,
             mblCtrl = ctrl,
             dissUsage = 'predictors',
             k = seq(40, 150, by = 10),
             method = 'gpr')

getPredictions(sbl.u)

resemble implements a function dedicated to non-linear modelling of complex visible and infrared spectral data based on memory-based learning (MBL, a.k.a instance-based learning or local modelling in the chemometrics literature). The package also includes functions for: computing and evaluate spectral similarity/dissimilarity matrices; projecting the spectra onto low dimensional orthogonal variables; removing irrelevant spectra from a reference set; etc.

The functions for computing and evaluate spectral similarity/dissimilarity matrices can be summarized as follows:

fDiss: Euclidean and Mahalanobis distances as well as the cosine dissimilarity (a.k.a spectral angle mapper)
corDiss: correlation and moving window correlation dissimilarity
sid: spectral information divergence between spectra or between the probability distributions of spectra
orthoDiss: principal components and partial least squares dissimilarity (including several options)
simEval: evaluates a given similarity/dissimilarity matrix based on the concept of side information

The functions for projecting the spectra onto low dimensional orthogonal variables are:

pcProjection: projects the spectra onto a principal component space
plsProjection: projects the spectra onto a partial least squares component space (a.k.a projection to latent structures)
orthoProjection: reproduces either the pcProjection or the plsProjection functions

The projection functions also offer different options for optimizing/selecting the number of components involved in the projection.

The functions modelling the spectra using memory-based learning are:

mblControl: controls some modelling aspects of the mbl function
mbl: models the spectra by memory-based learning

Some additional miscellaneous functions are:

print.mbl: prints a summary of the results obtained by the mbl function
plot.mbl: plots a summary of the results obtained by the mbl function
print.localOrthoDiss: prints local distance matrices generated with the orthoDiss function

In order to expand a little bit more the explanation on the mbl function, let's define first the basic input datasets:

• Reference (training) set: Dataset with n reference samples (e.g. spectral library) to be used in the calibration of spectral models. Xr represents the matrix of samples (containing the spectral predictor variables) and Yr represents a given response variable corresponding to Xr.

• Prediction set : Data set with m samples where the response variable (Yu) is unknown. However it can be predicted by applying a spectral model (calibrated by using Xr and Yr) on the spectra of these samples (Xu).

In order to predict each value in Yu, the mbl function takes each sample in Xu and searches in Xr for its k-nearest neighbours (most spectrally similar samples). Then a (local) model is calibrated with these (reference) neighbours and it immediately predicts the correspondent value in Yu from Xu. In the function, the k-nearest neighbour search is performed by computing spectral similarity/dissimilarity matrices between samples. The mbl function offers the following regression options for calibrating the (local) models:

'gpr': Gaussian process with linear kernel
'pls': Partial least squares
'wapls1': Weighted average partial least squares 1
'wapls2': Weighted average partial least squares 2 (no longer supported)

Keywords

• Infrared spectroscopy
• Chemometrics
• Local modelling
• Spectral library
• Lazy learning
• Soil spectroscopy

Other R'elated stuff

• Check our other project called prospectr.
• Check this presentation in which we used the resemble package to predict soil attributes from large scale soil spectral libraries.

Bug report and development version

You can send an e-mail to the package maintainer (ramirez.lopez.leo@gmail.com) or create an issue on github.