README.md

@@ -211,7 +211,15 @@ Release artifacts are accessible from bintray with following configuration (see
 
 ```kotlin
 repositories {
+    jcenter()
+    maven("https://clojars.org/repo")
+    maven("https://dl.bintray.com/egor-bogomolov/astminer/")
+    maven("https://dl.bintray.com/hotkeytlt/maven")
+    maven("https://dl.bintray.com/kotlin/kotlin-eap")
+    maven("https://dl.bintray.com/kotlin/kotlinx")
     maven("https://dl.bintray.com/mipt-npm/kscience")
+    maven("https://jitpack.io")
+    mavenCentral()
 }
 
 dependencies {
@@ -228,7 +236,15 @@ Development builds are uploaded to the separate repository:
 
 ```kotlin
 repositories {
+    jcenter()
+    maven("https://clojars.org/repo")
+    maven("https://dl.bintray.com/egor-bogomolov/astminer/")
+    maven("https://dl.bintray.com/hotkeytlt/maven")
+    maven("https://dl.bintray.com/kotlin/kotlin-eap")
+    maven("https://dl.bintray.com/kotlin/kotlinx")
     maven("https://dl.bintray.com/mipt-npm/dev")
+    maven("https://jitpack.io")
+    mavenCentral()
 }
 ```
 

build.gradle.kts

@@ -1,3 +1,5 @@
+import ru.mipt.npm.gradle.KSciencePublishPlugin
+
 plugins {
     id("ru.mipt.npm.project")
 }
@@ -17,6 +19,7 @@ allprojects {
         maven("https://dl.bintray.com/mipt-npm/dev")
         maven("https://dl.bintray.com/mipt-npm/kscience")
         maven("https://jitpack.io")
+        maven("http://logicrunch.research.it.uu.se/maven/")
         mavenCentral()
     }
 
@@ -25,7 +28,7 @@ allprojects {
 }
 
 subprojects {
-    if (name.startsWith("kmath")) apply<ru.mipt.npm.gradle.KSciencePublishPlugin>()
+    if (name.startsWith("kmath")) apply<KSciencePublishPlugin>()
 }
 
 readme {

examples/build.gradle.kts

@@ -10,6 +10,20 @@ plugins {
 allOpen.annotation("org.openjdk.jmh.annotations.State")
 sourceSets.register("benchmarks")
 
+repositories {
+    jcenter()
+    maven("https://clojars.org/repo")
+    maven("https://dl.bintray.com/egor-bogomolov/astminer/")
+    maven("https://dl.bintray.com/hotkeytlt/maven")
+    maven("https://dl.bintray.com/kotlin/kotlin-eap")
+    maven("https://dl.bintray.com/kotlin/kotlinx")
+    maven("https://dl.bintray.com/mipt-npm/dev")
+    maven("https://dl.bintray.com/mipt-npm/kscience")
+    maven("https://jitpack.io")
+    maven("http://logicrunch.research.it.uu.se/maven/")
+    mavenCentral()
+}
+
 dependencies {
     implementation(project(":kmath-ast"))
     implementation(project(":kmath-kotlingrad"))

examples/src/main/kotlin/kscience/kmath/ast/KotlingradSupport.kt

@@ -1,9 +1,9 @@
 package kscience.kmath.ast
 
 import kscience.kmath.asm.compile
+import kscience.kmath.expressions.derivative
 import kscience.kmath.expressions.invoke
 import kscience.kmath.expressions.symbol
-import kscience.kmath.kotlingrad.derivative
 import kscience.kmath.kotlingrad.differentiable
 import kscience.kmath.operations.RealField
 

kmath-commons/src/main/kotlin/kscience/kmath/commons/optimization/CMOptimizationProblem.kt

@@ -19,9 +19,8 @@ import kotlin.reflect.KClass
 public operator fun PointValuePair.component1(): DoubleArray = point
 public operator fun PointValuePair.component2(): Double = value
 
-public class CMOptimizationProblem(
+public class CMOptimizationProblem(override val symbols: List<Symbol>, ) :
-    override val symbols: List<Symbol>,
+    OptimizationProblem<Double>, SymbolIndexer, OptimizationFeature {
-) : OptimizationProblem<Double>, SymbolIndexer, OptimizationFeature {
     private val optimizationData: HashMap<KClass<out OptimizationData>, OptimizationData> = HashMap()
     private var optimizatorBuilder: (() -> MultivariateOptimizer)? = null
     public var convergenceChecker: ConvergenceChecker<PointValuePair> = SimpleValueChecker(DEFAULT_RELATIVE_TOLERANCE,
@@ -49,7 +48,7 @@ public class CMOptimizationProblem(
         addOptimizationData(objectiveFunction)
     }
 
-    public override fun diffExpression(expression: DifferentiableExpression<Double>): Unit {
+    public override fun diffExpression(expression: DifferentiableExpression<Double, Expression<Double>>) {
         expression(expression)
         val gradientFunction = ObjectiveFunctionGradient {
             val args = it.toMap()

kmath-commons/src/main/kotlin/kscience/kmath/commons/optimization/cmFit.kt

@@ -12,7 +12,6 @@ import kscience.kmath.structures.asBuffer
 import org.apache.commons.math3.analysis.differentiation.DerivativeStructure
 import org.apache.commons.math3.optim.nonlinear.scalar.GoalType
 
-
 /**
  * Generate a chi squared expression from given x-y-sigma data and inline model. Provides automatic differentiation
  */
@@ -21,7 +20,7 @@ public fun Fitting.chiSquared(
     y: Buffer<Double>,
     yErr: Buffer<Double>,
     model: DerivativeStructureField.(x: DerivativeStructure) -> DerivativeStructure,
-): DifferentiableExpression<Double> = chiSquared(DerivativeStructureField, x, y, yErr, model)
+): DifferentiableExpression<Double, Expression<Double>> = chiSquared(DerivativeStructureField, x, y, yErr, model)
 
 /**
  * Generate a chi squared expression from given x-y-sigma data and inline model. Provides automatic differentiation
@@ -31,7 +30,7 @@ public fun Fitting.chiSquared(
     y: Iterable<Double>,
     yErr: Iterable<Double>,
     model: DerivativeStructureField.(x: DerivativeStructure) -> DerivativeStructure,
-): DifferentiableExpression<Double> = chiSquared(
+): DifferentiableExpression<Double, Expression<Double>> = chiSquared(
     DerivativeStructureField,
     x.toList().asBuffer(),
     y.toList().asBuffer(),
@@ -39,7 +38,6 @@ public fun Fitting.chiSquared(
     model
 )
 
-
 /**
  * Optimize expression without derivatives
  */
@@ -48,16 +46,15 @@ public fun Expression<Double>.optimize(
     configuration: CMOptimizationProblem.() -> Unit,
 ): OptimizationResult<Double> = optimizeWith(CMOptimizationProblem, symbols = symbols, configuration)
 
-
 /**
  * Optimize differentiable expression
  */
-public fun DifferentiableExpression<Double>.optimize(
+public fun DifferentiableExpression<Double, Expression<Double>>.optimize(
     vararg symbols: Symbol,
     configuration: CMOptimizationProblem.() -> Unit,
 ): OptimizationResult<Double> = optimizeWith(CMOptimizationProblem, symbols = symbols, configuration)
 
-public fun DifferentiableExpression<Double>.minimize(
+public fun DifferentiableExpression<Double, Expression<Double>>.minimize(
     vararg startPoint: Pair<Symbol, Double>,
     configuration: CMOptimizationProblem.() -> Unit = {},
 ): OptimizationResult<Double> {

kmath-commons/src/test/kotlin/kscience/kmath/commons/optimization/OptimizeTest.kt

@@ -47,14 +47,17 @@ internal class OptimizeTest {
         val sigma = 1.0
         val generator = Distribution.normal(0.0, sigma)
         val chain = generator.sample(RandomGenerator.default(112667))
-        val x = (1..100).map { it.toDouble() }
+        val x = (1..100).map(Int::toDouble)
-        val y = x.map { it ->
+
+        val y = x.map {
             it.pow(2) + it + 1 + chain.nextDouble()
         }
-        val yErr = x.map { sigma }
+
-        val chi2 = Fitting.chiSquared(x, y, yErr) { x ->
+        val yErr = List(x.size) { sigma }
+
+        val chi2 = Fitting.chiSquared(x, y, yErr) { x1 ->
             val cWithDefault = bindOrNull(c) ?: one
-            bind(a) * x.pow(2) + bind(b) * x + cWithDefault
+            bind(a) * x1.pow(2) + bind(b) * x1 + cWithDefault
         }
 
         val result = chi2.minimize(a to 1.5, b to 0.9, c to 1.0)

kmath-core/src/commonMain/kotlin/kscience/kmath/expressions/DifferentiableExpression.kt

@@ -6,7 +6,7 @@ package kscience.kmath.expressions
  * @param T the type this expression takes as argument and returns.
  * @param R the type of expression this expression can be differentiated to.
  */
-public interface DifferentiableExpression<T, R : Expression<T>> : Expression<T> {
+public interface DifferentiableExpression<T, out R : Expression<T>> : Expression<T> {
     /**
      * Differentiates this expression by ordered collection of [symbols].
      *
@@ -43,6 +43,6 @@ public abstract class FirstDerivativeExpression<T, R : Expression<T>> : Differen
 /**
  * A factory that converts an expression in autodiff variables to a [DifferentiableExpression]
  */
-public fun interface AutoDiffProcessor<T : Any, I : Any, A : ExpressionAlgebra<T, I>, R : Expression<T>> {
+public fun interface AutoDiffProcessor<T : Any, I : Any, A : ExpressionAlgebra<T, I>, out R : Expression<T>> {
     public fun process(function: A.() -> I): DifferentiableExpression<T, R>
 }

kmath-kotlingrad/src/test/kotlin/kscience/kmath/kotlingrad/AdaptingTests.kt

@@ -13,13 +13,11 @@ import kotlin.test.assertTrue
 import kotlin.test.fail
 
 internal class AdaptingTests {
-    private val proto: DReal = DoublePrecision.prototype
-
     @Test
     fun symbol() {
         val c1 = MstAlgebra.symbol("x")
-        assertTrue(c1.toSVar(proto).name == "x")
+        assertTrue(c1.toSVar<KMathNumber<Double, RealField>>().name == "x")
-        val c2 = "kitten".parseMath().toSFun(proto)
+        val c2 = "kitten".parseMath().toSFun<KMathNumber<Double, RealField>>()
         if (c2 is SVar) assertTrue(c2.name == "kitten") else fail()
     }
 
@@ -27,15 +25,15 @@ internal class AdaptingTests {
     fun number() {
         val c1 = MstAlgebra.number(12354324)
         assertTrue(c1.toSConst<DReal>().doubleValue == 12354324.0)
-        val c2 = "0.234".parseMath().toSFun(proto)
+        val c2 = "0.234".parseMath().toSFun<KMathNumber<Double, RealField>>()
         if (c2 is SConst) assertTrue(c2.doubleValue == 0.234) else fail()
-        val c3 = "1e-3".parseMath().toSFun(proto)
+        val c3 = "1e-3".parseMath().toSFun<KMathNumber<Double, RealField>>()
         if (c3 is SConst) assertEquals(0.001, c3.value) else fail()
     }
 
     @Test
     fun simpleFunctionShape() {
-        val linear = "2*x+16".parseMath().toSFun(proto)
+        val linear = "2*x+16".parseMath().toSFun<KMathNumber<Double, RealField>>()
         if (linear !is Sum) fail()
         if (linear.left !is Prod) fail()
         if (linear.right !is SConst) fail()
@@ -43,8 +41,8 @@ internal class AdaptingTests {
 
     @Test
     fun simpleFunctionDerivative() {
-        val x = MstAlgebra.symbol("x").toSVar(proto)
+        val x = MstAlgebra.symbol("x").toSVar<KMathNumber<Double, RealField>>()
-        val quadratic = "x^2-4*x-44".parseMath().toSFun(proto)
+        val quadratic = "x^2-4*x-44".parseMath().toSFun<KMathNumber<Double, RealField>>()
         val actualDerivative = MstExpression(RealField, quadratic.d(x).toMst()).compile()
         val expectedDerivative = MstExpression(RealField, "2*x-4".parseMath()).compile()
         assertEquals(actualDerivative("x" to 123.0), expectedDerivative("x" to 123.0))
@@ -52,8 +50,8 @@ internal class AdaptingTests {
 
     @Test
     fun moreComplexDerivative() {
-        val x = MstAlgebra.symbol("x").toSVar(proto)
+        val x = MstAlgebra.symbol("x").toSVar<KMathNumber<Double, RealField>>()
-        val composition = "-sqrt(sin(x^2)-cos(x)^2-16*x)".parseMath().toSFun(proto)
+        val composition = "-sqrt(sin(x^2)-cos(x)^2-16*x)".parseMath().toSFun<KMathNumber<Double, RealField>>()
         val actualDerivative = MstExpression(RealField, composition.d(x).toMst()).compile()
 
         val expectedDerivative = MstExpression(

kmath-stat/src/commonMain/kotlin/kscience/kmath/stat/Fitting.kt

@@ -12,16 +12,18 @@ public object Fitting {
      * Generate a chi squared expression from given x-y-sigma data and inline model. Provides automatic differentiation
      */
     public fun <T : Any, I : Any, A> chiSquared(
-        autoDiff: AutoDiffProcessor<T, I, A>,
+        autoDiff: AutoDiffProcessor<T, I, A, Expression<T>>,
         x: Buffer<T>,
         y: Buffer<T>,
         yErr: Buffer<T>,
         model: A.(I) -> I,
-    ): DifferentiableExpression<T> where A : ExtendedField<I>, A : ExpressionAlgebra<T, I> {
+    ): DifferentiableExpression<T, Expression<T>> where A : ExtendedField<I>, A : ExpressionAlgebra<T, I> {
         require(x.size == y.size) { "X and y buffers should be of the same size" }
         require(y.size == yErr.size) { "Y and yErr buffer should of the same size" }
+
         return autoDiff.process {
             var sum = zero
+
             x.indices.forEach {
                 val xValue = const(x[it])
                 val yValue = const(y[it])
@@ -29,6 +31,7 @@ public object Fitting {
                 val modelValue = model(xValue)
                 sum += ((yValue - modelValue) / yErrValue).pow(2)
             }
+
             sum
         }
     }
@@ -45,6 +48,7 @@ public object Fitting {
     ): Expression<Double> {
         require(x.size == y.size) { "X and y buffers should be of the same size" }
         require(y.size == yErr.size) { "Y and yErr buffer should of the same size" }
+
         return Expression { arguments ->
             x.indices.sumByDouble {
                 val xValue = x[it]
@@ -56,4 +60,4 @@ public object Fitting {
             }
         }
     }
-}
+}

kmath-stat/src/commonMain/kotlin/kscience/kmath/stat/OptimizationProblem.kt

@@ -27,17 +27,17 @@ public interface OptimizationProblem<T : Any> {
     /**
      * Define the initial guess for the optimization problem
      */
-    public fun initialGuess(map: Map<Symbol, T>): Unit
+    public fun initialGuess(map: Map<Symbol, T>)
 
     /**
      * Set an objective function expression
      */
-    public fun expression(expression: Expression<T>): Unit
+    public fun expression(expression: Expression<T>)
 
     /**
      * Set a differentiable expression as objective function as function and gradient provider
      */
-    public fun diffExpression(expression: DifferentiableExpression<T>): Unit
+    public fun diffExpression(expression: DifferentiableExpression<T, Expression<T>>)
 
     /**
      * Update the problem from previous optimization run
@@ -50,9 +50,8 @@ public interface OptimizationProblem<T : Any> {
     public fun optimize(): OptimizationResult<T>
 }
 
-public interface OptimizationProblemFactory<T : Any, out P : OptimizationProblem<T>> {
+public fun interface OptimizationProblemFactory<T : Any, out P : OptimizationProblem<T>> {
     public fun build(symbols: List<Symbol>): P
-
 }
 
 public operator fun <T : Any, P : OptimizationProblem<T>> OptimizationProblemFactory<T, P>.invoke(
@@ -60,7 +59,6 @@ public operator fun <T : Any, P : OptimizationProblem<T>> OptimizationProblemFac
     block: P.() -> Unit,
 ): P = build(symbols).apply(block)
 
-
 /**
  * Optimize expression without derivatives using specific [OptimizationProblemFactory]
  */
@@ -78,7 +76,7 @@ public fun <T : Any, F : OptimizationProblem<T>> Expression<T>.optimizeWith(
 /**
  * Optimize differentiable expression using specific [OptimizationProblemFactory]
  */
-public fun <T : Any, F : OptimizationProblem<T>>  DifferentiableExpression<T>.optimizeWith(
+public fun <T : Any, F : OptimizationProblem<T>>  DifferentiableExpression<T, Expression<T>>.optimizeWith(
     factory: OptimizationProblemFactory<T, F>,
     vararg symbols: Symbol,
     configuration: F.() -> Unit,
@@ -88,4 +86,3 @@ public fun <T : Any, F : OptimizationProblem<T>>  DifferentiableExpression<T>.op
     problem.diffExpression(this)
     return problem.optimize()
 }
-

settings.gradle.kts

@@ -1,8 +1,7 @@
 pluginManagement {
     repositories {
-        mavenLocal()
-        jcenter()
         gradlePluginPortal()
+        jcenter()
         maven("https://dl.bintray.com/kotlin/kotlin-eap")
         maven("https://dl.bintray.com/mipt-npm/kscience")
         maven("https://dl.bintray.com/mipt-npm/dev")