Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/*
* Copyright 2015 University of Basel, Graphics and Vision Research Group
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package scalismo.statisticalmodel.dataset
import scalismo.common.{DiscreteDomain, DomainWarp, Vectorizer}
import scalismo.geometry.{_3D, EuclideanVector, NDSpace}
import scalismo.mesh.TriangleMesh
import scalismo.numerics.UniformMeshSampler3D
import scalismo.statisticalmodel.{GaussianProcess, LowRankGaussianProcess, PointDistributionModel, StatisticalMeshModel}
import scalismo.utils.Random
import scala.util.Try
/**
* Implements utility functions for evaluating the quality of a registered dataset
*/
object Crossvalidation {
type EvaluationFunction[D, DDomain[D] <: DiscreteDomain[D], A] =
(PointDistributionModel[D, DDomain], DDomain[D]) => A
/**
* Perform a leave one out crossvalidation. See nFoldCrossvalidation for details
*/
def leaveOneOutCrossvalidation[D, DDomain[D] <: DiscreteDomain[D], A](
dataCollection: DataCollection[D, DDomain, EuclideanVector[D]],
evalFun: EvaluationFunction[D, DDomain, A],
biasModelAndRank: Option[LowRankGaussianProcess[D, EuclideanVector[D]]] = None
)(
implicit
ndSpace: NDSpace[D],
domainWarp: DomainWarp[D, DDomain],
vectorizer: Vectorizer[EuclideanVector[D]],
rng: Random
) = {
nFoldCrossvalidation(dataCollection.size, dataCollection, evalFun, biasModelAndRank)
}
/**
* Perform an n-fold crossvalidation. Given the chosen number of folds, this method will repeatedly split the data collection
* into a training and and a test set.
* A [[StatisticalMeshModel]] is then built from the training set of each fold. In case a biasModel is provided, this model is always added to the model built from the training data.
*
* For each testing dataset in a fold, the evalFun is called to evaluate the quality of the model built from the training set.
*
* @return a sequence the size of the chosen number of folds that contains the sequence of evaluations for each data item in the fold's testing set,
* or an error if the model building for a fold failed.
*/
def nFoldCrossvalidation[D, DDomain[D] <: DiscreteDomain[D], A](
numFolds: Int,
dc: DataCollection[D, DDomain, EuclideanVector[D]],
evalFun: EvaluationFunction[D, DDomain, A],
biasModelAndRank: Option[LowRankGaussianProcess[D, EuclideanVector[D]]] = None
)(
implicit
ndSpace: NDSpace[D],
domainWarp: DomainWarp[D, DDomain],
vectorizer: Vectorizer[EuclideanVector[D]],
rng: Random
): Seq[Seq[A]] = {
val folds = dc.createCrossValidationFolds(numFolds)
val evalResultsForFolds = for (fold <- folds) yield {
val td = fold.trainingData
val pcaModel = PointDistributionModel.createUsingPCA(td)
val model = if (biasModelAndRank.isDefined) {
val biasModel = biasModelAndRank.get
PointDistributionModel.augmentModel[D, DDomain](pcaModel, biasModel)
} else {
pcaModel
}
val evalResults = for (testingItem <- fold.testingData.dataItems) yield {
val testMesh = domainWarp.transformWithField(pcaModel.reference, testingItem)
evalFun(model, testMesh)
}
evalResults
}
evalResultsForFolds
}
}
