rchemo0.1-3 package

Dimension Reduction, Regression and Discrimination for Chemometrics

aggmean

Centers of classes

aicplsr

AIC and Cp for Univariate PLSR Models

blocks

Block autoscaling

cglsr

CG Least Squares Models

checkdupl

Duplicated rows in datasets

checkna

Find and count NA values in a dataset

covsel

CovSel

dderiv

Derivation by finite difference

detrend

Polynomial de-trend transformation

dfplsr_cg

Degrees of freedom of Univariate PLSR Models

dkplsr

Direct KPLSR Models

dkrr

Direct KRR Models

dmnorm

Multivariate normal probability density

dtagg

Summary statistics of data subsets

dummy

Table of dummy variables

epo

External parameter orthogonalization (EPO)

euclsq

Matrix of distances

fda

Factorial discriminant analysis

getknn

KNN selection

gram

Kernel functions

gridcv

Cross-validation

gridscore

Tuning of predictive models on a validation dataset

headm

Display of the first part of a data set

interpl

Resampling of spectra by interpolation methods

knnda

KNN-DA

knnr

KNN-R

kpca

KPCA

kplsr

KPLSR Models

kplsrda

KPLSR-DA models

krr

KRR (LS-SVMR)

krrda

KRR-DA models

lda

LDA and QDA

lmr

Linear regression models

lmrda

LMR-DA models

locw

Locally weighted models

lwplsda_agg

Aggregation of KNN-LWPLSDA models with different numbers of LVs

lwplsr_agg

Aggregation of KNN-LWPLSR models with different numbers of LVs

lwplsr

KNN-LWPLSR

lwplsrda

KNN-LWPLS-DA Models

matW

Between and within covariance matrices

mavg

Smoothing by moving average

mbplsda

multi-block PLSDA models

mbplsr_mbplsda_allsteps

MBPLSR or MBPLSDA analysis steps

mbplsr

multi-block PLSR algorithms

odis

Orthogonal distances from a PCA or PLS score space

orthog

Orthogonalization of a matrix to another matrix

pca

PCA algorithms

pinv

Moore-Penrose pseudo-inverse of a matrix

plotjit

Jittered plot

plotscore

Plotting errors rates

plotsp

Plotting spectra

plotxna

Plotting Missing Data in a Matrix

plotxy

2-d scatter plot

plsda_agg

PLSDA with aggregation of latent variables

plsda

PLSDA models

plsr_agg

PLSR with aggregation of latent variables

plsr_plsda_allsteps

PLSR or PLSDA analysis steps

plsr

PLSR algorithms

rmgap

Removing vertical gaps in spectra

rr

Linear Ridge Regression

rrda

RR-DA models

sampcla

Within-class sampling

sampdp

Duplex sampling

sampks

Kennard-Stone sampling

savgol

Savitzky-Golay smoothing

scordis

Score distances (SD) in a PCA or PLS score space

scores

Residuals and prediction error rates

segmkf

Segments for cross-validation

selwold

Heuristic selection of the dimension of a latent variable model with t...

snv

Standard normal variate transformation (SNV)

soplsda

Block dimension reduction by SO-PLS-DA

soplsr_soplsda_allsteps

SOPLSR or SOPLSDA analysis steps

soplsr

Block dimension reduction by SO-PLS

sourcedir

Source R functions in a directory

summ

Description of the quantitative variables of a data set

svm

SVM Regression and Discrimination

transform

Generic transform function

vip

Variable Importance in Projection (VIP)

wdist

Distance-based weights

xfit

Matrix fitting from a PCA or PLS model

Data exploration and prediction with focus on high dimensional data and chemometrics. The package was initially designed about partial least squares regression and discrimination models and variants, in particular locally weighted PLS models (LWPLS). Then, it has been expanded to many other methods for analyzing high dimensional data. The name 'rchemo' comes from the fact that the package is orientated to chemometrics, but most of the provided methods are fully generic to other domains. Functions such as transform(), predict(), coef() and summary() are available. Tuning the predictive models is facilitated by generic functions gridscore() (validation dataset) and gridcv() (cross-validation). Faster versions are also available for models based on latent variables (LVs) (gridscorelv() and gridcvlv()) and ridge regularization (gridscorelb() and gridcvlb()).