Merge pull request #31 from spectre-team/develop

Master release with wavelet denoising finished

appveyor.yml

@@ -13,7 +13,7 @@ environment:
 configuration:
   - Debug
   - Release
-version: 3.5.2.{build}
+version: 3.14.0.{build}
 
 init:
 - cmd: echo Project - %APPVEYOR_PROJECT_NAME%

scripts/restore_nugets.bat

@@ -5,9 +5,10 @@ if errorlevel 1 (
 	echo Connection to GMrukwaAppVeyorFeed failed.
 	EXIT /B 1
 )
-nuget restore >nul 2>&1
+nuget restore >nuget.log 2>&1
 if errorlevel 1 (
 	echo NuGet failed to restore packages.
+	type nuget.log
 	EXIT /B 1
 )
 cd ..

src/Spectre.libClassifier.Tests/Spectre.libClassifier.Tests.vcxproj

@@ -22,7 +22,7 @@
   <PropertyGroup Label="Globals">
     <VCProjectVersion>15.0</VCProjectVersion>
     <ProjectGuid>{B6B8D608-32E7-4F09-B91D-EA5832D74C07}</ProjectGuid>
-    <RootNamespace>SpectrelibClassifierTests</RootNamespace>
+    <RootNamespace>Spectre::libClassifierTests</RootNamespace>
     <WindowsTargetPlatformVersion>8.1</WindowsTargetPlatformVersion>
   </PropertyGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />

src/Spectre.libClassifier.Tests/SvmTest.cpp

@@ -22,6 +22,7 @@ limitations under the License.
 #include "Spectre.libClassifier/OpenCvDataset.h"
 #include "Spectre.libClassifier/Svm.h"
 #include "Spectre.libClassifier/UntrainedClassifierException.h"
+#include "Spectre.libClassifier/InsufficientDataException.h"
 
 namespace
 {
@@ -71,6 +72,28 @@ TEST_F(SvmTest, predicts_sign)
     }
 }
 
+TEST_F(SvmTest, predicts_small_data)
+{
+    const std::vector<Label> tempLabels = { 1 };
+    const std::vector<DataType> tempFeatures = {
+        1., 1.
+    };
+    OpenCvDataset tempData(tempFeatures, tempLabels);
+    Svm svm;
+    EXPECT_THROW(svm.Fit(tempData), InsufficientDataException);
+}
+
+TEST_F(SvmTest, fit_with_non_binary_labels_throw_not_a_binary_exception)
+{
+    const std::vector<Label> tempLabels = { 2 };
+    const std::vector<DataType> tempFeatures = {
+        1., 1.
+    };
+    OpenCvDataset tempData(tempFeatures, tempLabels);
+    Svm svm;
+    EXPECT_THROW(svm.Fit(tempData), NotABinaryLabelException);
+}
+
 TEST_F(SvmTest, get_support_vector_number_test)
 {
     Svm svm;

src/Spectre.libClassifier/IClassifier.h

@@ -21,6 +21,7 @@ limitations under the License.
 #include <vector>
 #include "Spectre.libDataset/IReadOnlyDataset.h"
 #include "Spectre.libClassifier/Types.h"
+#include "Spectre.libDataset/Empty.h"
 
 namespace spectre::supervised
 {

src/Spectre.libClassifier/InsufficientDataException.cpp

@@ -0,0 +1,25 @@
+/*
+* InsufficientDataException.cpp
+* Throws when data does not cover all classes.
+*
+Copyright 2018 Spectre Team
+
+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.
+*/
+
+#include "InsufficientDataException.h"
+
+namespace spectre::supervised::exception
+{
+InsufficientDataException::InsufficientDataException(Label missingClass): ExceptionBase("Dataset is missing " + std::to_string(missingClass) + " label") {}
+}

src/Spectre.libClassifier/InsufficientDataException.h

@@ -0,0 +1,38 @@
+/*
+* InsufficientDataException.h
+* Throws when data does not cover all classes.
+*
+Copyright 2018 Spectre Team
+
+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.
+*/
+
+#pragma once
+#include "Spectre.libException/ExceptionBase.h"
+#include "NotABinaryLabelException.h"
+
+namespace spectre::supervised::exception
+{
+/// <summary>
+/// Throws when data does not cover all classes.
+/// </summary>
+class InsufficientDataException final : public core::exception::ExceptionBase
+{
+public:
+    /// <summary>
+    /// Initializes a new instance of the <see cref="InsufficientDataException"/> class.
+    /// </summary>
+    /// <param name="actual">The missing class .</param>
+    explicit InsufficientDataException(Label missingClass);
+};
+}

src/Spectre.libClassifier/RandomSplitter.cpp

@@ -35,17 +35,11 @@ RandomSplitter::RandomSplitter(double trainingRate, Seed rngSeed)
     : m_trainingRate(trainingRate),
       m_Seed(rngSeed)
 {
-    if (trainingRate >= 0)
-    {
-    }
-    else
+    if (trainingRate < 0)
     {
         throw exception::NegativeTrainingRateException(trainingRate);
     }
-    if (trainingRate <= 1)
-    {
-    }
-    else
+    if (trainingRate > 1)
     {
         throw exception::ExcessiveTrainingRateException(trainingRate);
     }

src/Spectre.libClassifier/Spectre.libClassifier.vcxproj

@@ -143,6 +143,7 @@
   <ItemGroup>
     <ClCompile Include="EmptyOpenCvDatasetException.cpp" />
     <ClCompile Include="ExcessiveTrainingRateException.cpp" />
+    <ClCompile Include="InsufficientDataException.cpp" />
     <ClCompile Include="NegativeTrainingRateException.cpp" />
     <ClCompile Include="NotABinaryLabelException.cpp" />
     <ClCompile Include="ObservationExtractor.cpp" />
@@ -157,6 +158,7 @@
     <ClInclude Include="EmptyOpenCvDatasetException.h" />
     <ClInclude Include="ExcessiveTrainingRateException.h" />
     <ClInclude Include="IClassifier.h" />
+    <ClInclude Include="InsufficientDataException.h" />
     <ClInclude Include="NegativeTrainingRateException.h" />
     <ClInclude Include="NotABinaryLabelException.h" />
     <ClInclude Include="ObservationExtractor.h" />

src/Spectre.libClassifier/Spectre.libClassifier.vcxproj.filters

@@ -60,6 +60,9 @@
     <ClCompile Include="Svm.cpp">
       <Filter>Source Files\Classifiers</Filter>
     </ClCompile>
+    <ClCompile Include="InsufficientDataException.cpp">
+      <Filter>Source Files\Exceptions</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="OpenCvDataset.h">
@@ -107,6 +110,9 @@
     <ClInclude Include="NotABinaryLabelException.h">
       <Filter>Header Files\Exceptions</Filter>
     </ClInclude>
+    <ClInclude Include="InsufficientDataException.h">
+      <Filter>Header Files\Exceptions</Filter>
+    </ClInclude>
   </ItemGroup>
   <ItemGroup>
     <None Include="packages.config" />

src/Spectre.libClassifier/Svm.cpp

@@ -21,6 +21,7 @@ limitations under the License.
 #include "Spectre.libClassifier/OpenCvDataset.h"
 #include "Spectre.libClassifier/UnsupportedDatasetTypeException.h"
 #include "Spectre.libClassifier/UntrainedClassifierException.h"
+#include "InsufficientDataException.h"
 
 namespace spectre::supervised
 {
@@ -35,6 +36,22 @@ Svm::Svm(unsigned int iterationsLimit, double tolerance)
 
 void Svm::Fit(LabeledDataset dataset)
 {
+    auto sum = 0u;
+    const auto positive = 1u;
+    const auto negative = 0u;
+    for (size_t i = 0u; i < dataset.size(); ++i)
+    {
+        const auto label = dataset.GetSampleMetadata(i);
+        if (label != negative && label != positive)
+        {
+            throw exception::NotABinaryLabelException(label, i, "dataset");
+        }
+        sum += label;
+    }
+    if (sum == 0u || sum == dataset.size())
+    {
+        throw exception::InsufficientDataException(sum == 0u ? positive : negative);
+    }
     const auto& data = asSupported(dataset);
     m_Svm->clear();
     m_Svm->train(data.getMatData(), cv::ml::ROW_SAMPLE, data.getMatLabels());

src/Spectre.libClassifier/Types.h

@@ -21,7 +21,6 @@ limitations under the License.
 #include <random>
 #include <opencv2/core/mat.hpp>
 #include <span.h>
-#include "Spectre.libDataset/Empty.h"
 
 namespace spectre::supervised
 {

src/Spectre.libGaClassifier.Tests/AllLabelTypesIncludedConditionTest.cpp

@@ -0,0 +1,83 @@
+/*
+* AllLabelTypesIncludedConditionTest.cpp
+* Tests AllLabelTypesIncludedCondition class.
+*
+Copyright 2018 Spectre Team
+
+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.
+*/
+
+#include <gtest/gtest.h>
+#include "Spectre.libGaClassifier/AllLabelTypesIncludedCondition.h"
+
+namespace
+{
+using namespace spectre::supervised;
+
+TEST(AllLabelTypesIncludedConditionInitialization, initializes_with_one_parameter)
+{
+    const std::vector<spectre::supervised::Label> labels{ true, true, true, true, true };
+    EXPECT_NO_THROW(AllLabelTypesIncludedCondition condition(labels));
+}
+
+class AllLabelTypesIncludedConditionTest : public ::testing::Test
+{
+public:
+    AllLabelTypesIncludedConditionTest() {}
+protected:
+    const spectre::algorithm::genetic::Individual individual = spectre::algorithm::genetic::Individual({ true, true, true, true, false });
+    const std::vector<spectre::supervised::Label> allTrueLabels{ true, true, true, true, false };
+    const std::vector<spectre::supervised::Label> allFalseLabels{ false, false, false, false, true };
+    const std::vector<spectre::supervised::Label> trueFalseLabels{ true, false, true, true, false };
+    const std::vector<spectre::supervised::Label> nonBinalyLabels{ 0, 1, 2, 1, 1, 5, 3, 3, 9, 0, 0, 2, 1, 1, 5, 5, 1 };
+    const spectre::algorithm::genetic::Individual nonBinaryIndividualThree =
+        spectre::algorithm::genetic::Individual({ true, false, true, false, false, false, true, true, false, false, false, true, false, false, false, false, false});
+    const spectre::algorithm::genetic::Individual nonBinaryIndividualSix =
+        spectre::algorithm::genetic::Individual({ true, true, true, false, true, false, true, true, true, false, true, true, true, false, true, false, false });
+};
+
+TEST_F(AllLabelTypesIncludedConditionTest, return_true_for_correct_label)
+{
+    AllLabelTypesIncludedCondition condition(trueFalseLabels);
+    auto result = condition.checkCurrentCondition(individual);
+    EXPECT_TRUE(result);
+}
+
+TEST_F(AllLabelTypesIncludedConditionTest, return_false_for_only_true_label_individual)
+{
+    AllLabelTypesIncludedCondition condition(allTrueLabels);
+    auto result = condition.checkCurrentCondition(individual);
+    EXPECT_FALSE(result);
+}
+
+TEST_F(AllLabelTypesIncludedConditionTest, return_false_for_only_false_label_individual)
+{
+    AllLabelTypesIncludedCondition condition(allFalseLabels);
+    auto result = condition.checkCurrentCondition(individual);
+    EXPECT_FALSE(result);
+}
+
+TEST_F(AllLabelTypesIncludedConditionTest, return_true_for_non_binary_label)
+{
+    AllLabelTypesIncludedCondition condition(nonBinalyLabels);
+    auto result = condition.checkCurrentCondition(nonBinaryIndividualSix);
+    EXPECT_TRUE(result);
+}
+
+TEST_F(AllLabelTypesIncludedConditionTest, return_false_for_non_binary_label)
+{
+    AllLabelTypesIncludedCondition condition(nonBinalyLabels);
+    auto result = condition.checkCurrentCondition(nonBinaryIndividualThree);
+    EXPECT_FALSE(result);
+}
+}