MixtureMachine API

tailhq · Jun 22, 2017 · fd818b6 · fd818b6
1 parent 45adb7b
commit fd818b6
Show file tree

Hide file tree

Showing 5 changed files with 37 additions and 10 deletions.
diff --git a/dynaml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/modelpipe/MixturePipe.scala b/dynaml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/modelpipe/MixturePipe.scala
@@ -5,11 +5,15 @@ import breeze.stats.distributions.{ContinuousDistr, Moments}
 import io.github.mandar2812.dynaml.algebra.{PartitionedPSDMatrix, PartitionedVector}
 import io.github.mandar2812.dynaml.models.gp.AbstractGPRegressionModel
 import io.github.mandar2812.dynaml.models.stp.{AbstractSTPRegressionModel, MVStudentsTModel}
-import io.github.mandar2812.dynaml.models.{ContinuousProcessModel, GenContinuousMixtureModel, SecondOrderProcessModel, StochasticProcessMixtureModel}
+import io.github.mandar2812.dynaml.models.{
+ContinuousProcessModel, GenContinuousMixtureModel,
+SecondOrderProcessModel, StochasticProcessMixtureModel}
 import io.github.mandar2812.dynaml.optimization.GloballyOptimizable
 import io.github.mandar2812.dynaml.pipes.DataPipe2
 import io.github.mandar2812.dynaml.probability.{ContinuousRVWithDistr, MatrixTRV, MultGaussianPRV, MultStudentsTPRV}
-import io.github.mandar2812.dynaml.probability.distributions.{BlockedMultiVariateGaussian, BlockedMultivariateStudentsT, HasErrorBars, MatrixT}
+import io.github.mandar2812.dynaml.probability.distributions.{
+BlockedMultiVariateGaussian, BlockedMultivariateStudentsT,
+HasErrorBars, MatrixT}
 
 import scala.reflect.ClassTag
 

diff --git a/dynaml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/models/gp/WarpedGPModel.scala b/dynaml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/models/gp/WarpedGPModel.scala
@@ -15,10 +15,11 @@ import scala.reflect.ClassTag
 
 /**
   * ::Experimental::
-  * @author mandar date 02/01/2017.
   *
   * A warped Gaussian Process.
-  */
+  * @author mandar date 02/01/2017.
+  *
+  * */
 @Experimental
 class WarpedGPModel[T, I:ClassTag](p: AbstractGPRegressionModel[T, I])(
   warpingFunc: PushforwardMap[Double, Double, Double])(

diff --git a/...l/optimization/ProbGPMixtureMachine.scala → ...ynaml/optimization/GPMixtureMachine.scala b/...l/optimization/ProbGPMixtureMachine.scala → ...ynaml/optimization/GPMixtureMachine.scala
@@ -16,10 +16,12 @@ import scala.reflect.ClassTag
   * @tparam I The index set/input domain of the GP model.
   * @author mandar2812 date 15/06/2017.
   * */
-class ProbGPMixtureMachine[T, I: ClassTag](
+class GPMixtureMachine[T, I: ClassTag](
   model: AbstractGPRegressionModel[T, I]) extends
-  MixtureMachine[T, I, Double, PartitionedVector, PartitionedPSDMatrix, BlockedMultiVariateGaussian,
-    MultGaussianPRV, AbstractGPRegressionModel[T, I]](model) {
+  MixtureMachine[
+    T, I, Double, PartitionedVector, PartitionedPSDMatrix,
+    BlockedMultiVariateGaussian, MultGaussianPRV,
+    AbstractGPRegressionModel[T, I]](model) {
 
   val (kernelPipe, noisePipe) = (system.covariance.asPipe, system.noiseModel.asPipe)
 
@@ -33,7 +35,8 @@ class ProbGPMixtureMachine[T, I: ClassTag](
     (model_state: Map[String, Double]) =>
       AbstractGPRegressionModel(
         kernelPipe(model_state), noisePipe(model_state),
-        system.mean)(system.data, system.npoints))
+        system.mean)(system.data, system.npoints)
+  )
 
   override val mixturePipe = new GPMixturePipe[T, I]
 

diff --git a/...ml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/optimization/MixtureMachine.scala b/...ml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/optimization/MixtureMachine.scala
@@ -131,5 +131,24 @@ BaseProcess <: ContinuousProcessModel[T, I, Y, W1]
     )
   }
 
+}
 
+object MixtureMachine {
+
+  def apply[
+    T, I: ClassTag, Y, YDomain, YDomainVar,
+    BaseDistr <: ContinuousDistr[YDomain] with Moments[YDomain, YDomainVar] with HasErrorBars[YDomain],
+    W1 <: ContinuousRVWithDistr[YDomain, BaseDistr],
+    BaseProcess <: ContinuousProcessModel[T, I, Y, W1]
+      with SecondOrderProcessModel[T, I, Y, Double, DenseMatrix[Double], W1]
+      with GloballyOptimizable](model: BaseProcess)(
+    confModelPipe: DataPipe[Map[String, Double], BaseProcess],
+    mixtPipe: DataPipe2[Seq[BaseProcess], DenseVector[Double], GenContinuousMixtureModel[
+      T, I, Y, YDomain, YDomainVar,
+      BaseDistr, W1, BaseProcess]]) =
+    new MixtureMachine[T, I, Y, YDomain, YDomainVar, BaseDistr, W1, BaseProcess](model) {
+      override val confToModel = confModelPipe
+      override val mixturePipe = mixtPipe
+    }
+
 }
diff --git a/scripts/stochasticPriors.sc b/scripts/stochasticPriors.sc
@@ -5,7 +5,7 @@ import io.github.mandar2812.dynaml.models.bayes.{LinearTrendESGPrior, LinearTren
 import io.github.mandar2812.dynaml.probability._
 import com.quantifind.charts.Highcharts._
 import io.github.mandar2812.dynaml.analysis.implicits._
-import io.github.mandar2812.dynaml.optimization.ProbGPMixtureMachine
+import io.github.mandar2812.dynaml.optimization.GPMixtureMachine
 import io.github.mandar2812.dynaml.pipes.Encoder
 import io.github.mandar2812.dynaml.probability.distributions.UnivariateGaussian
 
@@ -63,7 +63,7 @@ val sgpModel = sgp_prior.posteriorModel(dataset)
 gp_prior.globalOptConfig_(Map("gridStep" -> "0.0", "gridSize" -> "1", "globalOpt" -> "GS", "policy" -> "GS"))
 val gpModel1 = gp_prior.posteriorModel(dataset)
 
-val mixt_machine = new ProbGPMixtureMachine(gpModel1).setPrior(hyp_prior).setGridSize(2).setStepSize(0.50).setLogScale(true).setMaxIterations(200).setNumSamples(3)
+val mixt_machine = new GPMixtureMachine(gpModel1).setPrior(hyp_prior).setGridSize(2).setStepSize(0.50).setLogScale(true).setMaxIterations(200).setNumSamples(3)
 
 val (mix_model, mixt_model_conf) = mixt_machine.optimize(gp_prior.covariance.effective_state ++ gp_prior.noiseCovariance.effective_state)