public void testDecisonListWithNoTestsReturnsDefaultValue()
{
DecisionList dlist = new DecisionList("Yes", "No");
DataSet ds = DataSetFactory.getRestaurantDataSet();
Assert.AreEqual("No", dlist.predict(ds.getExample(0)));
}
static void ensembleLearningDemo()
{
try
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
ICollection <DecisionTree> stumps = DecisionTree.getStumpsFor(ds, "Yes", "No");
ICollection <ILearner> learners = CollectionFactory.CreateQueue <ILearner>();
System.Console.WriteLine("\nStump Learners vote to decide in this algorithm");
foreach (object stump in stumps)
{
DecisionTree sl = (DecisionTree)stump;
StumpLearner stumpLearner = new StumpLearner(sl, "No");
learners.Add(stumpLearner);
}
AdaBoostLearner learner = new AdaBoostLearner(learners, ds);
learner.Train(ds);
var answer = learner.Predict(ds.getExample(0));
int[] result = learner.Test(ds);
System.Console.WriteLine("\nThis Ensemble Learner classifies the data set with "
+ result[0]
+ " successes"
+ " and "
+ result[1]
+ " failures");
System.Console.WriteLine("\n");
}
catch (Exception e)
{
throw e;
}
}
static void decisionListDemo()
{
try
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
DecisionListLearner learner = new DecisionListLearner("Yes", "No",
new DecisionListTestFactory());
learner.Train(ds);
System.Console.WriteLine("The Induced DecisionList is");
System.Console.WriteLine(learner.getDecisionList());
int[] result = learner.Test(ds);
System.Console.WriteLine("\nThis Decision List classifies the data set with "
+ result[0]
+ " successes"
+ " and "
+ result[1]
+ " failures");
System.Console.WriteLine("\n");
}
catch (Exception e)
{
System.Console.WriteLine("Decision ListDemo Failed");
throw e;
}
}
public void testDecisionList()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
List <DLTest> dlTests = new DLTestFactory()
.createDLTestsWithAttributeCount(ds, 1);
Assert.AreEqual(26, dlTests.Count);
}
public void testStumpCreationForDataSet()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
List <DecisionTree> dt = DecisionTree.getStumpsFor(ds, YES,
"Unable to classify");
Assert.AreEqual(26, dt.Count);
}
public void testDecisionList()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
ICollection <aima.net.learning.inductive.DecisionListTest> dlTests = new DecisionListTestFactory()
.createDLTestsWithAttributeCount(ds, 1);
Assert.AreEqual(26, dlTests.Size());
}
public void testNonDestructiveRemoveExample()
{
DataSet ds1 = DataSetFactory.getRestaurantDataSet();
DataSet ds2 = ds1.removeExample(ds1.getExample(0));
Assert.AreEqual(12, ds1.size());
Assert.AreEqual(11, ds2.size());
}
public void testDLTestMatchesEvenOnMismatchedTargetAttributeValue()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
DLTest test = new DLTest();
test.add("type", "French");
Assert.IsTrue(test.matches(e));
}
public void testDLTestMatchFailsOnMismatchedExample()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
DLTest test = new DLTest();
test.add("type", "Thai");
Assert.IsFalse(test.matches(e));
}
public void testDLTestMatchSucceedsWithMatchedExample()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
DLTest test = new DLTest();
test.add("type", "French");
Assert.IsTrue(test.matches(e));
}
public void testDLTestMatchFailsOnMismatchedExample()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
aima.net.learning.inductive.DecisionListTest test = new aima.net.learning.inductive.DecisionListTest();
test.add("type", "Thai");
Assert.IsFalse(test.matches(e));
}
public void testDLTestMatchSucceedsWithMatchedExample()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
aima.net.learning.inductive.DecisionListTest test = new aima.net.learning.inductive.DecisionListTest();
test.add("type", "French");
Assert.IsTrue(test.matches(e));
}
public void testGainCalculation()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
double gain = ds.calculateGainFor("patrons");
Assert.AreEqual(0.541, gain, 0.001);
gain = ds.calculateGainFor("type");
Assert.AreEqual(0.0, gain, 0.001);
}
public void testInducedDecisionTreeClassifiesRestaurantDataSetCorrectly()
{
DecisionTreeLearner learner = new DecisionTreeLearner(
createInducedRestaurantDecisionTree(), "Unable to clasify");
int[] results = learner.test(DataSetFactory.getRestaurantDataSet());
Assert.AreEqual(12, results[0]);
Assert.AreEqual(0, results[1]);
}
public void testInducedTreeClassifiesDataSetCorrectly()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
DecisionTreeLearner learner = new DecisionTreeLearner();
learner.train(ds);
int[] result = learner.test(ds);
Assert.AreEqual(12, result[0]);
Assert.AreEqual(0, result[1]);
}
public void testBasicDataSetInformationCalculation()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
double infoForTargetAttribute = ds.getInformationFor();// this should
// be the
// generic
// distribution
Assert.AreEqual(1.0, infoForTargetAttribute, 0.001);
}
public void testMajorityLearner()
{
MajorityLearner learner = new MajorityLearner();
DataSet ds = DataSetFactory.getRestaurantDataSet();
learner.train(ds);
int[] result = learner.test(ds);
Assert.AreEqual(6, result[0]);
Assert.AreEqual(6, result[1]);
}
public void testDLTestMatchesEvenOnMismatchedTargetAttributeValue()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Example e = ds.getExample(0);
aima.net.learning.inductive.DecisionListTest test = new aima.net.learning.inductive.DecisionListTest();
test.add("type", "French");
Assert.IsTrue(test.matches(e));
}
public void testStumpCreationForSpecifiedAttributeValuePair()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
List <String> unmatchedValues = new List <String>();
unmatchedValues.Add(NO);
DecisionTree dt = DecisionTree.getStumpFor(ds, "alternate", YES, YES,
unmatchedValues, NO);
Assert.IsNotNull(dt);
}
public void testCurrentBestLearnerOnRestaurantDataSet()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
CurrentBestLearner learner = new CurrentBestLearner("Yes");
learner.train(ds);
int[] result = learner.test(ds);
Assert.AreEqual(12, result[0]);
Assert.AreEqual(0, result[1]);
}
public void testDecisionListLearnerReturnsFailureWhenTestsEmpty()
{
// tests second base case of DL Learner
DecisionListLearner learner = new DecisionListLearner("Yes", "No",
new MockDLTestFactory(new List <DLTest>()));
DataSet ds = DataSetFactory.getRestaurantDataSet();
learner.train(ds);
Assert.AreEqual(DecisionListLearner.FAILURE, learner.predict(ds
.getExample(0)));
}
public void testDecisionListLearnerReturnsFailureWhenTestsEmpty()
{
// tests second base case of DL Learner
DecisionListLearner learner = new DecisionListLearner("Yes", "No", new MockDLTestFactory(CollectionFactory.CreateQueue <aima.net.learning.inductive.DecisionListTest>()));
DataSet ds = DataSetFactory.getRestaurantDataSet();
learner.Train(ds);
Assert.AreEqual(DecisionListLearner.FAILURE,
learner.Predict(ds.getExample(0)));
}
public void testDecisionListTestRunOnRestaurantDataSet()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
DecisionListLearner learner = new DecisionListLearner("Yes", "No",
new DLTestFactory());
learner.train(ds);
int[] result = learner.test(ds);
Assert.AreEqual(12, result[0]);
Assert.AreEqual(0, result[1]);
}
public void testDecisionListWithSingleTestReturnsTestValueIfTestSuccessful()
{
DecisionList dlist = new DecisionList("Yes", "No");
DataSet ds = DataSetFactory.getRestaurantDataSet();
DLTest test = new DLTest();
test.add("type", "French");
dlist.add(test, "test1success");
Assert.AreEqual("test1success", dlist.predict(ds.getExample(0)));
}
public void testDecisionListLearnerReturnsNegativeDLWhenDataSetEmpty()
{
// tests first base case of DL Learner
DecisionListLearner learner = new DecisionListLearner("Yes", "No",
new MockDLTestFactory(null));
DataSet ds = DataSetFactory.getRestaurantDataSet();
DataSet empty = ds.emptyDataSet();
learner.train(empty);
Assert.AreEqual("No", learner.predict(ds.getExample(0)));
Assert.AreEqual("No", learner.predict(ds.getExample(1)));
Assert.AreEqual("No", learner.predict(ds.getExample(2)));
}
public void testDefaultUsedWhenTrainingDataSetHasNoExamples()
{
// tests RecursionBaseCase#1
DataSet ds = DataSetFactory.getRestaurantDataSet();
DecisionTreeLearner learner = new DecisionTreeLearner();
DataSet ds2 = ds.emptyDataSet();
Assert.AreEqual(0, ds2.size());
learner.train(ds2);
Assert.AreEqual("Unable To Classify", learner.predict(ds
.getExample(0)));
}
public void testDLTestReturnsMatchedAndUnmatchedExamplesCorrectly()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
DLTest test = new DLTest();
test.add("type", "Burger");
DataSet matched = test.matchedExamples(ds);
Assert.AreEqual(4, matched.size());
DataSet unmatched = test.unmatchedExamples(ds);
Assert.AreEqual(8, unmatched.size());
}
public void testLoadsDatasetFile()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Assert.AreEqual(12, ds.size());
Example first = ds.getExample(0);
Assert.AreEqual(YES, first.getAttributeValueAsString("alternate"));
Assert.AreEqual("$$$", first.getAttributeValueAsString("price"));
Assert.AreEqual("0-10",
first.getAttributeValueAsString("wait_estimate"));
Assert.AreEqual(YES, first.getAttributeValueAsString("will_wait"));
Assert.AreEqual(YES, first.targetValue());
}
public void testDataSetSplit()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
Dictionary <String, DataSet> hash = ds.splitByAttribute("patrons");// this
// should
// be
// the
// generic
// distribution
Assert.AreEqual(3, hash.Keys.Count);
Assert.AreEqual(6, hash["Full"].size());
Assert.AreEqual(2, hash["None"].size());
Assert.AreEqual(4, hash["Some"].size());
}
public void testDataSetSplit()
{
DataSet ds = DataSetFactory.getRestaurantDataSet();
IMap <string, DataSet> hash = ds.splitByAttribute("patrons");// this
// should
// be
// the
// generic
// distribution
Assert.AreEqual(3, hash.GetKeys().Size());
Assert.AreEqual(6, hash.Get("Full").size());
Assert.AreEqual(2, hash.Get("None").size());
Assert.AreEqual(4, hash.Get("Some").size());
}
