Simpson and spline integration

This commit is contained in:
Alexander Nozik 2021-05-22 20:08:49 +03:00
parent 18509f1259
commit a24c8dcbce
12 changed files with 260 additions and 40 deletions

View File

@ -0,0 +1,45 @@
/*
* Copyright 2018-2021 KMath contributors.
* Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
*/
package space.kscience.kmath.functions
import space.kscience.kmath.interpolation.SplineInterpolator
import space.kscience.kmath.interpolation.interpolatePolynomials
import space.kscience.kmath.operations.DoubleField
import space.kscience.kmath.real.step
import space.kscience.kmath.structures.map
import space.kscience.plotly.Plotly
import space.kscience.plotly.UnstablePlotlyAPI
import space.kscience.plotly.makeFile
import space.kscience.plotly.models.functionXY
import space.kscience.plotly.scatter
@OptIn(UnstablePlotlyAPI::class)
fun main() {
val function: UnivariateFunction<Double> = { x ->
if (x in 30.0..50.0) {
1.0
} else {
0.0
}
}
val xs = 0.0..100.0 step 0.5
val ys = xs.map(function)
val polynomial: PiecewisePolynomial<Double> = SplineInterpolator.double.interpolatePolynomials(xs, ys)
val polyFunction = polynomial.asFunction(DoubleField, 0.0)
Plotly.plot {
scatter {
name = "interpolated"
functionXY(25.0..55.0, 0.1) { polyFunction(it) }
}
scatter {
name = "original"
functionXY(25.0..55.0, 0.1) { function(it) }
}
}.makeFile()
}

View File

@ -16,6 +16,8 @@ public fun <T> Ring<T>.sum(data: Iterable<T>): T = data.fold(zero) { left, right
add(left, right) add(left, right)
} }
//TODO replace by sumOf with multi-receivers
/** /**
* Returns the sum of all elements in the sequence in this [Ring]. * Returns the sum of all elements in the sequence in this [Ring].
* *

View File

@ -74,9 +74,9 @@ class NumberNDFieldTest {
val division = array1.combine(array2, Double::div) val division = array1.combine(array2, Double::div)
} }
object L2Norm : Norm<StructureND<out Number>, Double> { object L2Norm : Norm<StructureND<Number>, Double> {
@OptIn(PerformancePitfall::class) @OptIn(PerformancePitfall::class)
override fun norm(arg: StructureND<out Number>): Double = override fun norm(arg: StructureND<Number>): Double =
kotlin.math.sqrt(arg.elements().sumOf { it.second.toDouble() }) kotlin.math.sqrt(arg.elements().sumOf { it.second.toDouble() })
} }

View File

@ -69,18 +69,23 @@ public fun <T, A> Polynomial<T>.differentiate(
public fun <T, A> Polynomial<T>.integrate( public fun <T, A> Polynomial<T>.integrate(
algebra: A, algebra: A,
): Polynomial<T> where A : Field<T>, A : NumericAlgebra<T> = algebra { ): Polynomial<T> where A : Field<T>, A : NumericAlgebra<T> = algebra {
Polynomial(coefficients.mapIndexed { index, t -> t / number(index) }) val integratedCoefficients = buildList<T>(coefficients.size + 1) {
add(zero)
coefficients.forEachIndexed{ index, t -> add(t / (number(index) + one)) }
}
Polynomial(integratedCoefficients)
} }
/** /**
* Compute a definite integral of a given polynomial in a [range] * Compute a definite integral of a given polynomial in a [range]
*/ */
@UnstableKMathAPI @UnstableKMathAPI
public fun <T : Comparable<T>, A> Polynomial<T>.integrate( public fun <T : Comparable<T>> Polynomial<T>.integrate(
algebra: A, algebra: Field<T>,
range: ClosedRange<T>, range: ClosedRange<T>,
): T where A : Field<T>, A : NumericAlgebra<T> = algebra { ): T = algebra {
value(algebra, range.endInclusive) - value(algebra, range.start) val integral = integrate(algebra)
integral.value(algebra, range.endInclusive) - integral.value(algebra, range.start)
} }
/** /**

View File

@ -10,13 +10,7 @@ import space.kscience.kmath.structures.Buffer
import space.kscience.kmath.structures.asBuffer import space.kscience.kmath.structures.asBuffer
import space.kscience.kmath.structures.indices import space.kscience.kmath.structures.indices
/**
* Set of univariate integration ranges. First components correspond to ranges themselves, second components to number of
* integration nodes per range
*/
public class UnivariateIntegrandRanges(public val ranges: List<Pair<ClosedRange<Double>, Int>>) : IntegrandFeature {
public constructor(vararg pairs: Pair<ClosedRange<Double>, Int>) : this(pairs.toList())
}
/** /**
* A simple one-pass integrator based on Gauss rule * A simple one-pass integrator based on Gauss rule

View File

@ -6,36 +6,53 @@
package space.kscience.kmath.integration package space.kscience.kmath.integration
import space.kscience.kmath.operations.Field import space.kscience.kmath.operations.Field
import space.kscience.kmath.structures.Buffer import space.kscience.kmath.operations.invoke
import space.kscience.kmath.structures.BufferFactory import space.kscience.kmath.operations.sum
/** /**
* Use double pass Simpson rule integration with a fixed number of points.
* Requires [UnivariateIntegrandRanges] or [IntegrationRange] and [IntegrandMaxCalls]
* [IntegrationRange] - the univariate range of integration. By default uses 0..1 interval. * [IntegrationRange] - the univariate range of integration. By default uses 0..1 interval.
* [IntegrandMaxCalls] - the maximum number of function calls during integration. For non-iterative rules, always uses the maximum number of points. By default uses 10 points. * [IntegrandMaxCalls] - the maximum number of function calls during integration. For non-iterative rules, always uses the maximum number of points. By default uses 10 points.
*/ */
public class SimpsonIntegrator<T : Any>( public class SimpsonIntegrator<T : Any>(
public val algebra: Field<T>, public val algebra: Field<T>,
public val bufferFactory: BufferFactory<T>,
) : UnivariateIntegrator<T> { ) : UnivariateIntegrator<T> {
override fun integrate(integrand: UnivariateIntegrand<T>): UnivariateIntegrand<T> = with(algebra) {
val numPoints = integrand.getFeature<IntegrandMaxCalls>()?.maxCalls ?: 100 private fun integrateRange(
require(numPoints >= 4) integrand: UnivariateIntegrand<T>, range: ClosedRange<Double>, numPoints: Int,
val range = integrand.getFeature<IntegrationRange>()?.range ?: 0.0..1.0 ): T = algebra {
val h: Double = (range.endInclusive - range.start) / (numPoints - 1) val h: Double = (range.endInclusive - range.start) / (numPoints - 1)
val points: Buffer<T> = bufferFactory(numPoints) { i -> val values: List<T> = List(numPoints) { i ->
integrand.function(range.start + i * h) integrand.function(range.start + i * h)
}// equally distributed point }// equally distributed point
fun simpson(index: Int) = h / 3 * (points[index - 1] + 4 * points[index] + points[index + 1]) //TODO don't use list, reassign values instead
fun simpson(index: Int) = h / 3 * (values[index - 1] + 4 * values[index] + values[index + 1])
var res = zero var res = zero
res += simpson(1) / 1.5 //border points with 1.5 factor res += simpson(1) / 1.5 //border points with 1.5 factor
for (i in 2 until (points.size - 2)) { for (i in 2 until (values.size - 2)) {
//each half-interval is computed twice, therefore /2
res += simpson(i) / 2 res += simpson(i) / 2
} }
res += simpson(points.size - 2) / 1.5 //border points with 1.5 factor res += simpson(values.size - 2) / 1.5 //border points with 1.5 factor
return integrand + IntegrandValue(res) + IntegrandCallsPerformed(integrand.calls + points.size) return res
}
override fun integrate(integrand: UnivariateIntegrand<T>): UnivariateIntegrand<T> {
val ranges = integrand.getFeature<UnivariateIntegrandRanges>()
return if (ranges != null) {
val res = algebra.sum(ranges.ranges.map { integrateRange(integrand, it.first, it.second) })
integrand + IntegrandValue(res) + IntegrandCallsPerformed(integrand.calls + ranges.ranges.sumOf { it.second })
} else {
val numPoints = integrand.getFeature<IntegrandMaxCalls>()?.maxCalls ?: 100
require(numPoints >= 4) { "Simpson integrator requires at least 4 nodes" }
val range = integrand.getFeature<IntegrationRange>()?.range ?: 0.0..1.0
val res = integrateRange(integrand, range, numPoints)
integrand + IntegrandValue(res) + IntegrandCallsPerformed(integrand.calls + numPoints)
}
} }
} }
public inline val <reified T : Any> Field<T>.simpsonIntegrator: SimpsonIntegrator<T> public val <T : Any> Field<T>.simpsonIntegrator: SimpsonIntegrator<T> get() = SimpsonIntegrator(this)
get() = SimpsonIntegrator(this, Buffer.Companion::auto)

View File

@ -5,8 +5,98 @@
package space.kscience.kmath.integration package space.kscience.kmath.integration
public class SplineIntegrator<T>: UnivariateIntegrator<T> { import space.kscience.kmath.functions.PiecewisePolynomial
override fun integrate(integrand: UnivariateIntegrand<T>): UnivariateIntegrand<T> { import space.kscience.kmath.functions.integrate
TODO("Not yet implemented") import space.kscience.kmath.interpolation.PolynomialInterpolator
import space.kscience.kmath.interpolation.SplineInterpolator
import space.kscience.kmath.interpolation.interpolatePolynomials
import space.kscience.kmath.misc.UnstableKMathAPI
import space.kscience.kmath.operations.DoubleField
import space.kscience.kmath.operations.Field
import space.kscience.kmath.operations.invoke
import space.kscience.kmath.operations.sum
import space.kscience.kmath.structures.Buffer
import space.kscience.kmath.structures.DoubleBuffer
import space.kscience.kmath.structures.MutableBufferFactory
import space.kscience.kmath.structures.map
/**
* Compute analytical indefinite integral of this [PiecewisePolynomial], keeping all intervals intact
*/
@UnstableKMathAPI
public fun <T : Comparable<T>> PiecewisePolynomial<T>.integrate(algebra: Field<T>): PiecewisePolynomial<T> =
PiecewisePolynomial(pieces.map { it.first to it.second.integrate(algebra) })
/**
* Compute definite integral of given [PiecewisePolynomial] piece by piece in a given [range]
* Requires [UnivariateIntegrationNodes] or [IntegrationRange] and [IntegrandMaxCalls]
*/
@UnstableKMathAPI
public fun <T : Comparable<T>> PiecewisePolynomial<T>.integrate(
algebra: Field<T>, range: ClosedRange<T>,
): T = algebra.sum(
pieces.map { (region, poly) ->
val intersectedRange = maxOf(range.start, region.start)..minOf(range.endInclusive, region.endInclusive)
//Check if polynomial range is not used
if (intersectedRange.start == intersectedRange.endInclusive) algebra.zero
else poly.integrate(algebra, intersectedRange)
}
)
/**
* A generic spline-interpolation-based analytic integration
* [IntegrationRange] - the univariate range of integration. By default uses 0..1 interval.
* [IntegrandMaxCalls] - the maximum number of function calls during integration. For non-iterative rules, always uses the maximum number of points. By default uses 10 points.
*/
@UnstableKMathAPI
public class SplineIntegrator<T : Comparable<T>>(
public val algebra: Field<T>,
public val bufferFactory: MutableBufferFactory<T>,
) : UnivariateIntegrator<T> {
override fun integrate(integrand: UnivariateIntegrand<T>): UnivariateIntegrand<T> = algebra {
val range = integrand.getFeature<IntegrationRange>()?.range ?: 0.0..1.0
val interpolator: PolynomialInterpolator<T> = SplineInterpolator(algebra, bufferFactory)
val nodes: Buffer<Double> = integrand.getFeature<UnivariateIntegrationNodes>()?.nodes ?: run {
val numPoints = integrand.getFeature<IntegrandMaxCalls>()?.maxCalls ?: 100
val step = (range.endInclusive - range.start) / (numPoints - 1)
DoubleBuffer(numPoints) { i -> range.start + i * step }
}
val values = nodes.map(bufferFactory) { integrand.function(it) }
val polynomials = interpolator.interpolatePolynomials(
nodes.map(bufferFactory) { number(it) },
values
)
val res = polynomials.integrate(algebra, number(range.start)..number(range.endInclusive))
integrand + IntegrandValue(res) + IntegrandCallsPerformed(integrand.calls + nodes.size)
} }
} }
/**
* A simplified double-based spline-interpolation-based analytic integration
* [IntegrationRange] - the univariate range of integration. By default uses 0..1 interval.
* [IntegrandMaxCalls] - the maximum number of function calls during integration. For non-iterative rules, always uses the maximum number of points. By default uses 10 points.
*/
@UnstableKMathAPI
public object DoubleSplineIntegrator : UnivariateIntegrator<Double> {
override fun integrate(integrand: UnivariateIntegrand<Double>): UnivariateIntegrand<Double> {
val range = integrand.getFeature<IntegrationRange>()?.range ?: 0.0..1.0
val interpolator: PolynomialInterpolator<Double> = SplineInterpolator(DoubleField, ::DoubleBuffer)
val nodes: Buffer<Double> = integrand.getFeature<UnivariateIntegrationNodes>()?.nodes ?: run {
val numPoints = integrand.getFeature<IntegrandMaxCalls>()?.maxCalls ?: 100
val step = (range.endInclusive - range.start) / (numPoints - 1)
DoubleBuffer(numPoints) { i -> range.start + i * step }
}
val values = nodes.map { integrand.function(it) }
val polynomials = interpolator.interpolatePolynomials(nodes, values)
val res = polynomials.integrate(DoubleField, range)
return integrand + IntegrandValue(res) + IntegrandCallsPerformed(integrand.calls + nodes.size)
}
}
@UnstableKMathAPI
public inline val DoubleField.splineIntegrator: UnivariateIntegrator<Double>
get() = DoubleSplineIntegrator

View File

@ -6,6 +6,8 @@
package space.kscience.kmath.integration package space.kscience.kmath.integration
import space.kscience.kmath.misc.UnstableKMathAPI import space.kscience.kmath.misc.UnstableKMathAPI
import space.kscience.kmath.structures.Buffer
import space.kscience.kmath.structures.DoubleBuffer
import kotlin.jvm.JvmInline import kotlin.jvm.JvmInline
import kotlin.reflect.KClass import kotlin.reflect.KClass
@ -35,6 +37,19 @@ public typealias UnivariateIntegrator<T> = Integrator<UnivariateIntegrand<T>>
@JvmInline @JvmInline
public value class IntegrationRange(public val range: ClosedRange<Double>) : IntegrandFeature public value class IntegrationRange(public val range: ClosedRange<Double>) : IntegrandFeature
/**
* Set of univariate integration ranges. First components correspond to ranges themselves, second components to number of
* integration nodes per range
*/
public class UnivariateIntegrandRanges(public val ranges: List<Pair<ClosedRange<Double>, Int>>) : IntegrandFeature {
public constructor(vararg pairs: Pair<ClosedRange<Double>, Int>) : this(pairs.toList())
}
public class UnivariateIntegrationNodes(public val nodes: Buffer<Double>) : IntegrandFeature {
public constructor(vararg nodes: Double) : this(DoubleBuffer(nodes))
}
/** /**
* Value of the integrand if it is present or null * Value of the integrand if it is present or null
*/ */

View File

@ -15,10 +15,16 @@ import space.kscience.kmath.operations.Ring
import space.kscience.kmath.structures.Buffer import space.kscience.kmath.structures.Buffer
import space.kscience.kmath.structures.asBuffer import space.kscience.kmath.structures.asBuffer
/**
* And interpolator for data with x column type [X], y column type [Y].
*/
public fun interface Interpolator<T, X : T, Y : T> { public fun interface Interpolator<T, X : T, Y : T> {
public fun interpolate(points: XYColumnarData<T, X, Y>): (X) -> Y public fun interpolate(points: XYColumnarData<T, X, Y>): (X) -> Y
} }
/**
* And interpolator returning [PiecewisePolynomial] function
*/
public interface PolynomialInterpolator<T : Comparable<T>> : Interpolator<T, T, T> { public interface PolynomialInterpolator<T : Comparable<T>> : Interpolator<T, T, T> {
public val algebra: Ring<T> public val algebra: Ring<T>

View File

@ -24,7 +24,7 @@ class SimpsonIntegralTest {
@Test @Test
fun gaussUniform() { fun gaussUniform() {
val res = DoubleField.simpsonIntegrator.integrate(35.0..100.0) { x -> val res = DoubleField.simpsonIntegrator.integrate(35.0..100.0, IntegrandMaxCalls(20)) { x ->
if (x in 30.0..50.0) { if (x in 30.0..50.0) {
1.0 1.0
} else { } else {
@ -33,6 +33,4 @@ class SimpsonIntegralTest {
} }
assertEquals(15.0, res.value, 0.5) assertEquals(15.0, res.value, 0.5)
} }
} }

View File

@ -0,0 +1,48 @@
/*
* Copyright 2018-2021 KMath contributors.
* Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
*/
package space.kscience.kmath.integration
import space.kscience.kmath.functions.Polynomial
import space.kscience.kmath.functions.integrate
import space.kscience.kmath.misc.UnstableKMathAPI
import space.kscience.kmath.operations.DoubleField
import kotlin.math.PI
import kotlin.math.sin
import kotlin.test.Test
import kotlin.test.assertEquals
@OptIn(UnstableKMathAPI::class)
class SplineIntegralTest {
@Test
fun integratePolynomial(){
val polynomial = Polynomial(1.0, 2.0, 3.0)
val integral = polynomial.integrate(DoubleField,1.0..2.0)
assertEquals(11.0, integral, 0.001)
}
@Test
fun gaussSin() {
val res = DoubleField.splineIntegrator.integrate(0.0..2 * PI, IntegrandMaxCalls(5)) { x ->
sin(x)
}
assertEquals(0.0, res.value, 1e-2)
}
@Test
fun gaussUniform() {
val res = DoubleField.splineIntegrator.integrate(35.0..100.0, IntegrandMaxCalls(20)) { x ->
if(x in 30.0..50.0){
1.0
} else {
0.0
}
}
assertEquals(15.0, res.value, 0.5)
}
}

View File

@ -1,11 +1,11 @@
pluginManagement { pluginManagement {
repositories { repositories {
maven("https://repo.kotlin.link")
mavenCentral() mavenCentral()
gradlePluginPortal() gradlePluginPortal()
maven("https://repo.kotlin.link")
} }
val toolsVersion = "0.9.8" val toolsVersion = "0.9.9"
val kotlinVersion = "1.5.0" val kotlinVersion = "1.5.0"
plugins { plugins {
@ -15,8 +15,8 @@ pluginManagement {
kotlin("multiplatform") version kotlinVersion kotlin("multiplatform") version kotlinVersion
kotlin("jvm") version kotlinVersion kotlin("jvm") version kotlinVersion
kotlin("plugin.allopen") version kotlinVersion kotlin("plugin.allopen") version kotlinVersion
id("org.jetbrains.kotlinx.benchmark") version "0.3.0" id("org.jetbrains.kotlinx.benchmark") version "0.3.1"
kotlin("jupyter.api") version "0.9.1-61" kotlin("jupyter.api") version "0.10.0-25"
} }
} }