Add proper test for symmetric matrices eigenValueDecomposition

This commit is contained in:
Alexander Nozik 2024-08-18 22:45:33 +03:00
parent b818a8981f
commit 1619a49017
10 changed files with 77 additions and 19 deletions

View File

@ -0,0 +1,40 @@
/*
* Copyright 2018-2024 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.linear
import space.kscience.kmath.commons.linear.CMLinearSpace
import space.kscience.kmath.ejml.EjmlLinearSpaceDDRM
import space.kscience.kmath.nd.StructureND
import space.kscience.kmath.operations.algebra
import space.kscience.kmath.structures.Float64
import kotlin.random.Random
fun main() {
val dim = 46
val random = Random(123)
val u = Float64.algebra.linearSpace.buildMatrix(dim, dim) { i, j -> if (i <= j) random.nextDouble() else 0.0 }
listOf(CMLinearSpace, EjmlLinearSpaceDDRM).forEach { algebra ->
with(algebra) {
//create a simmetric matrix
val matrix = buildMatrix(dim, dim) { row, col ->
if (row >= col) u[row, col] else u[col, row]
}
val eigen = matrix.getOrComputeAttribute(EIG) ?: error("Failed to compute eigenvalue decomposition")
check(
StructureND.contentEquals(
matrix,
eigen.v dot eigen.d dot eigen.v.transposed(),
1e-4
)
) { "$algebra decomposition failed" }
println("$algebra eigenvalue decomposition complete and checked" )
}
}
}

View File

@ -7,6 +7,7 @@ package space.kscience.kmath.linear
import space.kscience.attributes.PolymorphicAttribute
import space.kscience.attributes.safeTypeOf
import space.kscience.kmath.UnstableKMathAPI
public interface EigenDecomposition<T> {
/**
@ -24,5 +25,6 @@ public class EigenDecompositionAttribute<T> :
PolymorphicAttribute<EigenDecomposition<T>>(safeTypeOf()),
MatrixAttribute<EigenDecomposition<T>>
@UnstableKMathAPI
public val <T> MatrixScope<T>.EIG: EigenDecompositionAttribute<T>
get() = EigenDecompositionAttribute()

View File

@ -1,13 +1,20 @@
plugins {
id("space.kscience.gradle.jvm")
id("space.kscience.gradle.mpp")
}
val ejmlVerision = "0.43.1"
dependencies {
api(projects.kmathCore)
api(projects.kmathComplex)
api("org.ejml:ejml-all:$ejmlVerision")
kscience {
jvm()
jvmMain {
api(projects.kmathCore)
api(projects.kmathComplex)
api("org.ejml:ejml-all:$ejmlVerision")
}
jvmTest {
implementation(projects.testUtils)
}
}
readme {
@ -29,11 +36,3 @@ readme {
ref = "src/main/kotlin/space/kscience/kmath/ejml/EjmlLinearSpace.kt"
) { "LinearSpace implementations." }
}
//kotlin.sourceSets.main {
// val codegen by tasks.creating {
// ejmlCodegen(kotlin.srcDirs.first().absolutePath + "/space/kscience/kmath/ejml/_generated.kt")
// }
//
// kotlin.srcDirs(files().builtBy(codegen))
//}

View File

@ -278,8 +278,8 @@ public object EjmlLinearSpaceDDRM : EjmlLinearSpace<Double, Float64Field, DMatri
}
override val v: Matrix<Float64> by lazy {
val eigenvectors = List(origin.numRows) { cmEigen.getEigenVector(it) }
buildMatrix(origin.numRows, origin.numCols) { row, column ->
val eigenvectors = List(origin.numRows) { cmEigen.getEigenVector(it) }.filterNotNull()
buildMatrix(eigenvectors.size, origin.numCols) { row, column ->
eigenvectors[row][column]
}
}

View File

@ -18,12 +18,15 @@ import space.kscience.kmath.nd.StructureND
import space.kscience.kmath.nd.toArray
import space.kscience.kmath.operations.algebra
import space.kscience.kmath.structures.Float64
import space.kscience.kmath.testutils.assertStructureEquals
import kotlin.random.Random
import kotlin.random.asJavaRandom
import kotlin.test.*
internal fun <T : Any> assertMatrixEquals(expected: StructureND<T>, actual: StructureND<T>) {
assertTrue { StructureND.contentEquals(expected, actual) }
expected.elements().forEach { (index, value) ->
assertEquals(value, actual[index], "Structure element with index ${index.toList()} should be equal to $value but is ${actual[index]}")
}
}
@OptIn(UnstableKMathAPI::class)
@ -108,8 +111,12 @@ internal class EjmlMatrixTest {
@Test
fun eigenValueDecomposition() = EjmlLinearSpaceDDRM {
val matrix = EjmlDoubleMatrix(randomMatrix)
val dim = 46
val u = buildMatrix(dim, dim) { i, j -> if (i <= j) random.nextDouble() else 0.0 }
val matrix = buildMatrix(dim, dim) { row, col ->
if (row >= col) u[row, col] else u[col, row]
}
val eigen = matrix.getOrComputeAttribute(EIG) ?: fail()
assertMatrixEquals(matrix, eigen.v dot eigen.d dot eigen.v.transposed())
assertStructureEquals(matrix, eigen.v dot eigen.d dot eigen.v.transposed())
}
}

View File

@ -5,6 +5,8 @@
package space.kscience.kmath.testutils
import space.kscience.kmath.PerformancePitfall
import space.kscience.kmath.nd.StructureND
import space.kscience.kmath.structures.Buffer
import space.kscience.kmath.structures.Float64
import space.kscience.kmath.structures.indices
@ -19,3 +21,11 @@ public fun assertBufferEquals(expected: Buffer<Float64>, result: Buffer<Float64>
assertEquals(expected[it], result[it], tolerance)
}
}
@OptIn(PerformancePitfall::class)
public fun assertStructureEquals(expected: StructureND<Float64>, result: StructureND<Float64>, tolerance: Double = 1e-4) {
assertEquals(expected.shape, result.shape, "Structure shape mismatch")
expected.indices.forEach {
assertEquals(expected[it], result[it], tolerance)
}
}