コード例 #1
0
def runExperiment():
    with open("data_banknote_authentication.txt", "r") as dataset:
        bankNoteData = removeAttrsRandom(np.asarray([line.split(",") for line in dataset.readlines()], dtype=np.float64))
        numFolds = 10
        crossVal = CrossValidation(bankNoteData, numFolds)
        for fold in range(crossVal.folds):
            print("\nRunning Fold : " + str(fold + 1))
            trainingData = crossVal.getTrainingData(fold)
            testingData = crossVal.getTestingData(fold)
            attrIndices = [i for i in range(bankNoteData.shape[1] - 1)]
            decisionTree = DecisionTree(bankNoteData, trainingData, attrIndices)
            makeDecisionTree(bankNoteData, decisionTree.root, decisionTree.root.allowedAttrs, None)
            outputs = [int(bankNoteData[item, 4]) for item in testingData]
            crossVal.addOutputs(outputs)
            queries = [bankNoteData[item, :4] for item in testingData]
            predictedOutputs = [classifyQuery([query], decisionTree.root) for query in queries]
            crossVal.addPredictedOutputs(predictedOutputs)
            print("OUTPUTS = " + str(outputs))
            print("PREDICTED OUTPUTS = " + str(predictedOutputs))
        
        crossVal.printConfusionMatrix()
        
    print("ACCURACY  = " + str(crossVal.getAverageAccuracy() * 100) + "%")
    print("PRECISION = " + str(crossVal.getAveragePrecision() * 100) + "%")
    print("RECALL    = " + str(crossVal.getAverageRecall() * 100) + "%")
コード例 #2
0
def runExperiment():
    with open("data_banknote_authentication.txt", "r") as dataset:
        bankNoteData = np.asarray([line.split(",") for line in dataset.readlines()], dtype=np.float64)

        numFolds = 10
        crossVal = CrossValidation(bankNoteData, numFolds)
        for fold in range(crossVal.folds):
            print("\nRunning Fold : " + str(fold + 1))
            trainingData = crossVal.getTrainingData(fold)
            testingData = crossVal.getTestingData(fold)
            outputs = [int(bankNoteData[item, 4]) for item in testingData]
            crossVal.addOutputs(outputs)
            randomForest = buildRandomForest(bankNoteData, trainingData)
            queries = [bankNoteData[item, :4] for item in testingData]
            predictedOutputs = [classifyQuery(randomForest, query) for query in queries]
            crossVal.addPredictedOutputs(predictedOutputs)
            print("OUTPUTS = " + str(outputs))
            print("PREDICTED OUTPUTS = " + str(predictedOutputs))

        crossVal.printConfusionMatrix()

    print("ACCURACY  = " + str(crossVal.getAverageAccuracy() * 100) + "%")
    print("PRECISION = " + str(crossVal.getAveragePrecision() * 100) + "%")
    print("RECALL    = " + str(crossVal.getAverageRecall() * 100) + "%")