simulateSpectra-class: Create a list with a simulated data set of spectra

In asmitapoddar/BayesSentinel: Bayes Classification of Spetrometric Data

Description

Simulate one or more Gaussian spectra at regularly sampling time

Usage

simulateSpectra(...)

Arguments

...	any paramaters to be input into the function

Slots

nbPixel

number of pixels belonging to class k

nbCluster

number of cluster

nbSpectrum

number of spectra

simulationType

type of simulation. Available options are "gaussian" and "tstudent". Default is "gaussian".

modelname

type of model to be used to build covariance matrix. Available options are "full" and "parsimonious". Default is "full".

kernelSpectra

type of kernel to be used to simulate spectra. Available options are "diag", "epanechnikov", "gaussian", "exponential", "uniform", "quadratic" , "circular", "triangular", "rational quadratic", "inverse multiquadratic". Default is "gaussian".

kernelTime

type of kernel to be used for simulating time. Available options are "diag", "epanechnikov", "gaussian", "exponential", "uniform", "quadratic", "circular", "triangular", "rational quadratic", "inverse multiquadratic". Default is "gaussian".

sigma

a vector of size nbSpectrum giving the variance level of the spectrum

nbSampling

number of time intervals of the simulation

times

time intervals of the simulation

width

the width of the kernel to use for "gaussian" simulation. Default is 50.

gamma

degrees of freedom used for simulating "tstudent" distribution of data. Default is 3.

labels

class labels of the data

result

return a list of simulated data

Author(s)

Serge Iovleff, Asmita Poddar & Florent Latimier

Examples

m = new("simulateSpectra")
res = simulate(m)

asmitapoddar/BayesSentinel documentation built on May 10, 2019, 1:18 a.m.