get confusion matrix

from sklearn.model_selection import train_test_split

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix

from sklearn.model_selection import GridSearchCV

from sklearn.linear_model import Perceptron, SGDClassifier

from sklearn.neural_network import MLPClassifier

    return accuracy, precision, recall, f1

def getConfusionMatrix():

    if len(X) == 0:

        preprocessing()

    if tree == None:

        loadModels()

    results = {'Perceptron': (confusion_matrix(Y, pla.predict(X)) / len(Y)).tolist(),

               'Stochastic Gradient Descent': (confusion_matrix(Y, sgd.predict(X)) / len(Y)).tolist(),

               'Neural Network': (confusion_matrix(Y, nn.predict(X)) / len(Y)).tolist(),

               'Decision Tree': (confusion_matrix(Y, tree.predict(X)) / len(Y)).tolist()}

    return results

# partial fit the new prediction data to each algorithm

def partialFitNewData(message, label):

    before = getPerformanceMetrics()

    return val

@app.route('/getConfusionMatrix', methods=["GET"])

def getConfusionMatrix():

    return ml.getConfusionMatrix();

# Run the app!

if __name__ == '__main__':

    app.run(port = 8080)