Commit f003d9e1 authored by Larry Schultheis's avatar Larry Schultheis
Browse files

Auto stash before merge of "development" and "origin/development"

parent 2e5e1d38

ml.py

+40 −0
Original line number Diff line number Diff line
@@ -21,6 +21,46 @@ sgd = None 
nn = None

tree = None



def getPerformanceMetrics(y_actual, y_pred):

    if len(X) == 0:

        preprocessing()

    

    if tree == None:

        loadModels()

        

    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=0)



    pla_yPred = pla.predict(X_test)

    sgd_yPred = sgd.predict(X_test)

    nn_yPred = nn.predict(X_test)

    tree_yPred = tree.predict(X_test)

    

    accuracy = [accuracy_score(Y_test, pla_yPred), 

                  accuracy_score(Y_test, sgd_yPred),

                  accuracy_score(Y_test, nn_yPred),

                  accuracy_score(Y_test, tree_yPred)]

    

    precision = [precision_score(Y_test, pla_yPred, average='macro'),

                  precision_score(Y_test, sgd_yPred, average='macro'),

                  precision_score(Y_test, nn_yPred, average='macro'),

                  precision_score(Y_test, tree_yPred, average='macro')]

        

    recall = [recall_score(Y_test, pla_yPred, average='macro'),

              recall_score(Y_test, sgd_yPred, average='macro'),

              recall_score(Y_test, nn_yPred, average='macro'),

              recall_score(Y_test, tree_yPred, average='macro')]

    

    f1 = [recall_score(Y_test, pla_yPred, average='macro'),

          recall_score(Y_test, sgd_yPred, average='macro'),

          recall_score(Y_test, nn_yPred, average='macro'),

          recall_score(Y_test, tree_yPred, average='macro')]

    

    # accuracy = accuracy_score(y_actual, y_pred)

    # precision = precision_score(y_actual, y_pred, average='macro')

    # recall = recall_score(y_actual, y_pred, average='macro')

    # f1 = f1_score(y_actual, y_pred, average='macro')

    

    return accuracy, precision, recall, f1



def printMetrics(y_actual, y_pred):

    print('Accuracy: ', accuracy_score(y_actual, y_pred))

server.py

+8 −5
Original line number Diff line number Diff line
@@ -50,11 +50,14 @@ def correctPrediction(): 
# TODO make this get the actual ranking of algorithms

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

def getTopAlgorithms():

    name = ["This (in use)", "Comes", "From", "The", "Server"]

    f1 = ["5", "4", "3", "2", "1"]

    precision = ["1", "2", "3", "4", "5"]

    recall = ["1", "2", "3", "4", "5"]

    accuracy = ["1", "2", "3", "4", "5"]

    # name = ["This (in use)", "Comes", "From", "The", "Server"]

    # f1 = ["5", "4", "3", "2", "1"]

    # precision = ["1", "2", "3", "4", "5"]

    # recall = ["1", "2", "3", "4", "5"]

    # accuracy = ["1", "2", "3", "4", "5"]

    

    name = "Perceptron", "Stochastic Gradient Descent", "Neural Network", "Decision Tree"

    accuracy, precision, recall, f1 = ml.getPerformanceMetrics()



    val = {

        "name": name,

web/js/main.js

+32 −32
Original line number Diff line number Diff line
@@ -109,38 +109,38 @@ function getTopAlgorithms() { 
            replyObj = JSON.parse(replyString);

            console.log(replyObj)

            // put the algorithm names in their ranked order

            document.getElementById("alg1_name").innerHTML = "<b>" + replyObj.name[0] + "</b>";

            document.getElementById("alg2_name").innerText = replyObj.name[1];

            document.getElementById("alg3_name").innerText = replyObj.name[2];

            document.getElementById("alg4_name").innerText = replyObj.name[3];

            document.getElementById("alg5_name").innerText = replyObj.name[4];



            // put the algorithms' F1 scores in their ranked order

            document.getElementById("alg1_f1").innerHTML = "<b>" + replyObj.f1[0] + "</b>";

            document.getElementById("alg2_f1").innerText = replyObj.f1[1];

            document.getElementById("alg3_f1").innerText = replyObj.f1[2];

            document.getElementById("alg4_f1").innerText = replyObj.f1[3];

            document.getElementById("alg5_f1").innerText = replyObj.f1[4];



            document.getElementById("alg1_acc").innerHTML = "<b>" + replyObj.accuracy[0] + "</b>";

            document.getElementById("alg2_acc").innerText = replyObj.accuracy[1];

            document.getElementById("alg3_acc").innerText = replyObj.accuracy[2];

            document.getElementById("alg4_acc").innerText = replyObj.accuracy[3];

            document.getElementById("alg5_acc").innerText = replyObj.accuracy[4];



            document.getElementById("alg1_precision").innerHTML = "<b>" + replyObj.precision[0] + "</b>";

            document.getElementById("alg2_precision").innerText = replyObj.precision[1];

            document.getElementById("alg3_precision").innerText = replyObj.precision[2];

            document.getElementById("alg4_precision").innerText = replyObj.precision[3];

            document.getElementById("alg5_precision").innerText = replyObj.precision[4];



            document.getElementById("alg1_recall").innerHTML = "<b>" + replyObj.recall[0] + "</b>";

            document.getElementById("alg2_recall").innerText = replyObj.recall[1];

            document.getElementById("alg3_recall").innerText = replyObj.recall[2];

            document.getElementById("alg4_recall").innerText = replyObj.recall[3];

            document.getElementById("alg5_recall").innerText = replyObj.recall[4];                        



            loadAlgGraph(replyObj)

            // document.getElementById("alg1_name").innerHTML = "<b>" + replyObj.name[0] + "</b>";

            // document.getElementById("alg2_name").innerText = replyObj.name[1];

            // document.getElementById("alg3_name").innerText = replyObj.name[2];

            // document.getElementById("alg4_name").innerText = replyObj.name[3];

            // document.getElementById("alg5_name").innerText = replyObj.name[4];



            // // put the algorithms' F1 scores in their ranked order

            // document.getElementById("alg1_f1").innerHTML = "<b>" + replyObj.f1[0] + "</b>";

            // document.getElementById("alg2_f1").innerText = replyObj.f1[1];

            // document.getElementById("alg3_f1").innerText = replyObj.f1[2];

            // document.getElementById("alg4_f1").innerText = replyObj.f1[3];

            // document.getElementById("alg5_f1").innerText = replyObj.f1[4];



            // document.getElementById("alg1_acc").innerHTML = "<b>" + replyObj.accuracy[0] + "</b>";

            // document.getElementById("alg2_acc").innerText = replyObj.accuracy[1];

            // document.getElementById("alg3_acc").innerText = replyObj.accuracy[2];

            // document.getElementById("alg4_acc").innerText = replyObj.accuracy[3];

            // document.getElementById("alg5_acc").innerText = replyObj.accuracy[4];



            // document.getElementById("alg1_precision").innerHTML = "<b>" + replyObj.precision[0] + "</b>";

            // document.getElementById("alg2_precision").innerText = replyObj.precision[1];

            // document.getElementById("alg3_precision").innerText = replyObj.precision[2];

            // document.getElementById("alg4_precision").innerText = replyObj.precision[3];

            // document.getElementById("alg5_precision").innerText = replyObj.precision[4];



            // document.getElementById("alg1_recall").innerHTML = "<b>" + replyObj.recall[0] + "</b>";

            // document.getElementById("alg2_recall").innerText = replyObj.recall[1];

            // document.getElementById("alg3_recall").innerText = replyObj.recall[2];

            // document.getElementById("alg4_recall").innerText = replyObj.recall[3];

            // document.getElementById("alg5_recall").innerText = replyObj.recall[4];                        



            // loadAlgGraph(replyObj)

        }

    }

    http.send();