• Example Search
• Project Search

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

• doddle-model-master
  • .github
    • PULL_REQUEST_TEMPLATE.md
    • ISSUE_TEMPLATE
      • feature_request.md
      • bug_report.md
    • CONTRIBUTING.md
  • src
    • main
      • resources
        • datasets
          • iris.csv
          • dummy_csv_reading.csv
          • breast_cancer.csv
          • boston_housing_prices.csv
          • high_school_test.csv
      • scala
        • io
          • picnicml
            • doddlemodel
              • package.scala
              • modelselection
                • GroupKFoldSplitter.scala
                • modelselection.scala
                • CrossValidation.scala
                • DataSplitter.scala
                • KFoldSplitter.scala
                • HyperparameterSearch.scala
              • typeclasses
                • Classifier.scala
                • Predictor.scala
                • Transformer.scala
                • Estimator.scala
                • Regressor.scala
              • dummy
                • classification
                  • StratifiedClassifier.scala
                  • MostFrequentClassifier.scala
                  • UniformClassifier.scala
                • regression
                  • MedianRegressor.scala
                  • MeanRegressor.scala
              • impute
                • MeanValueImputer.scala
                • MostFrequentValueImputer.scala
              • syntax
                • ClassifierSyntax.scala
                • RegressorSyntax.scala
                • TransformerSyntax.scala
                • PredictorSyntax.scala
                • OptionSyntax.scala
              • linear
                • typeclasses
                  • LinearModel.scala
                  • LinearClassifier.scala
                  • LinearRegressor.scala
                • PoissonRegression.scala
                • LogisticRegression.scala
                • SoftmaxClassifier.scala
                • LinearRegression.scala
              • data
                • ResourceDatasetLoaders.scala
                • package.scala
                • DatasetUtils.scala
                • CsvLoader.scala
                • Feature.scala
                • data.scala
              • preprocessing
                • Binarizer.scala
                • OneHotEncoder.scala
                • RangeScaler.scala
                • Normalizer.scala
                • Norms.scala
                • StandardScaler.scala
              • pipeline
                • package.scala
                • Pipeline.scala
              • metrics
                • ClassificationMetrics.scala
                • package.scala
                • RankingMetrics.scala
                • Metric.scala
                • RegressionMetrics.scala
      • scala-2.13+
        • io
          • picnicml
            • doddlemodel
              • CrossScalaCompat.scala
      • scala-2.12-
        • io
          • picnicml
            • doddlemodel
              • CrossScalaCompat.scala
    • test
      • scala
        • io
          • picnicml
            • doddlemodel
              • modelselection
                • GroupKFoldSplitterTest.scala
                • KFoldSplitterTest.scala
                • HyperparameterSearchTest.scala
              • integration
                • LogisticRegressionTest.scala
                • LinearRegressionTest.scala
                • SoftmaxClassifierTest.scala
                • PoissonRegressionTest.scala
              • TestingUtils.scala
              • dummy
                • classification
                  • StratifiedClassifierTest.scala
                  • MostFrequentClassifierTest.scala
                • regression
                  • MeanRegressorTest.scala
                  • MedianRegressorTest.scala
              • impute
                • MostFrequentValueImputerTest.scala
                • MeanValueImputerTest.scala
              • syntax
                • SyntaxTest.scala
              • linear
                • LinearRegressorTest.scala
                • LogisticRegressionTest.scala
                • LinearRegressionTest.scala
                • SoftmaxClassifierTest.scala
                • PoissonRegressionTest.scala
                • LinearClassifierTest.scala
              • data
                • ResourceDatasetLoadersTest.scala
                • DatasetUtilsTest.scala
                • CsvLoaderTest.scala
                • FeatureTest.scala
              • preprocessing
                • NormsTest.scala
                • BinarizerTest.scala
                • NormalizerTest.scala
                • OneHotEncoderTest.scala
                • StandardScalerTest.scala
                • RangeScalerTest.scala
              • pipeline
                • PipelineTest.scala
              • metrics
                • RankingMetricsTest.scala
                • ClassificationMetricsTest.scala
                • RegressionMetricsTest.scala
  • .circleci
    • config.yml
  • LICENSE
  • project
    • Dependencies.scala
    • build.properties
    • Version.scala
    • plugins.sbt
  • README.md
  • build.sbt
  • .gitignore

package io.picnicml.doddlemodel.linear

import breeze.linalg.{DenseMatrix, DenseVector}
import io.picnicml.doddlemodel.TestingUtils
import io.picnicml.doddlemodel.data.{Features, RealVector, Target}
import io.picnicml.doddlemodel.linear.SoftmaxClassifier.ev
import org.scalactic.{Equality, TolerantNumerics}
import org.scalatest.flatspec.AnyFlatSpec
import org.scalatest.matchers.should.Matchers

class SoftmaxClassifierTest extends AnyFlatSpec with Matchers with TestingUtils {

  implicit val tolerance: Equality[Float] = TolerantNumerics.tolerantFloatEquality(1e-3f)

  "Softmax classifier" should "calculate the value of the loss function" in {
    val w = DenseVector(1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 1.0f)
    val x = DenseMatrix(
      List(3.0f, 1.0f, 2.0f),
      List(-1.0f, -2.0f, 2.0f),
      List(-2.0f, 1.0f, 0.0f)
    )
    val y = DenseVector(1.0f, 0.0f, 2.0f)

    val model = ev.copy(SoftmaxClassifier(lambda = 1.0f), numClasses = 3)
    ev.lossStateless(model, w, x, y) shouldEqual 19.843778223530194f
  }

  it should "calculate the gradient of the loss function wrt. to model parameters" in {
    for (_ <- 1 to 1000) {
      val w = DenseVector.rand[Float](5 * 9, rand = randomUniform)
      val x = DenseMatrix.rand[Float](10, 5, rand = randomUniform)
      val y = DenseVector.rangeF(0, 10)
      testGrad(w, x, y)
    }

    def testGrad(w: RealVector, x: Features, y: Target) = {
      val model = ev.copy(SoftmaxClassifier(lambda = 0.5f), numClasses = 10)
      breezeEqual(
        gradApprox(w => ev.lossStateless(model, w, x, y), w),
        ev.lossGradStateless(model, w, x, y)
      ) shouldEqual true
    }
  }

  it should "prevent the usage of negative L2 regularization strength" in {
    an [IllegalArgumentException] shouldBe thrownBy(SoftmaxClassifier(lambda = -0.5f))
  }
}