def testGenerate(self): generate = DecisionTreeF.generate() self.X[:, 15:25] = self.X[:, 15:25]*100 decisionTree = generate() decisionTree.setWaveletInds(numpy.arange(100)) decisionTree.learnModel(self.X, self.y) self.assertEquals(numpy.intersect1d(numpy.arange(15,25), decisionTree.getFeatureInds()).shape[0], 10) predY = decisionTree.predict(self.X) #Now test when all features are wavelets decisionTree = generate() decisionTree.learnModel(self.X, self.y) self.assertEquals(numpy.intersect1d(numpy.arange(15,25), decisionTree.getFeatureInds()).shape[0], 10) predY = decisionTree.predict(self.X)
def testGenerate(self): generate = DecisionTreeF.generate() self.X[:, 15:25] = self.X[:, 15:25] * 100 decisionTree = generate() decisionTree.setWaveletInds(numpy.arange(100)) decisionTree.learnModel(self.X, self.y) self.assertEquals( numpy.intersect1d(numpy.arange(15, 25), decisionTree.getFeatureInds()).shape[0], 10) predY = decisionTree.predict(self.X) #Now test when all features are wavelets decisionTree = generate() decisionTree.learnModel(self.X, self.y) self.assertEquals( numpy.intersect1d(numpy.arange(15, 25), decisionTree.getFeatureInds()).shape[0], 10) predY = decisionTree.predict(self.X)
def testSetWeight(self): learner = DecisionTreeF() learner.setWeight(0.8)