Package edu.cnu.mdi.splot.fit

Class ValueSampler

java.lang.Object

edu.cnu.mdi.splot.fit.ValueSampler

public final class ValueSampler
extends Object

Utility methods for sampling Evaluator functions for plotting or further analysis.

Method Summary

Modifier and Type	Method	Description
static double[]	sampleAt(Evaluator f, double[] xs)	Sample a function at explicitly supplied x values.
static double[][]	sampleUniform(Evaluator f, double xmin, double xmax, int n)	Sample a function uniformly over [xmin, xmax].
static double[][]	sampleUniformFinite(Evaluator f, double xmin, double xmax, int n)	Sample uniformly, skipping non-finite results (NaN/Inf).

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details

sampleUniform

public static double[][] sampleUniform(Evaluator f,
 double xmin,
 double xmax,
 int n)

Sample a function uniformly over [xmin, xmax].

Parameters:

f - function to evaluate

xmin - minimum x

xmax - maximum x

n - number of sample points (n >= 2)

Returns:

array { xs, ys } each of length n

sampleAt

public static double[] sampleAt(Evaluator f,
 double[] xs)

Sample a function at explicitly supplied x values.

Parameters:

f - function to evaluate

xs - x values

Returns:

y values (same length as xs)

sampleUniformFinite

public static double[][] sampleUniformFinite(Evaluator f,
 double xmin,
 double xmax,
 int n)

Sample uniformly, skipping non-finite results (NaN/Inf). Useful when plotting functions that may blow up at the edges.

Parameters:

f - function

xmin - minimum x

xmax - maximum x

n - nominal number of samples

Returns:

array { xs, ys } with only finite points retained