def classfy():
# Split the data into columns and read
warnings.warn("Variables are collinear.")
datainput = pd.read_csv("trainingset.csv")
# Set the outcome and dedlete it
y = datainput['State']
del datainput['State']
# Split data into Test & Training set where test data is 30% & raining data is 70%
x_train, x_test, y_train, y_test = train_test_split(datainput,
y,
test_size=0.3)
# Next use Bayesian Classifier
classify3 = BernoulliNB()
# Train the model
classify3.fit(x_train, y_train)
# Use the model on the test data
predicted3 = classify3.predict(x_test)
print("NB", predicted3)
nb = metrics.accuracy_score(y_test, predicted3) * 100
print("The accuracy score using the Naive Bayes Classifier is ->")
print(metrics.accuracy_score(y_test, predicted3))
print('---------------------------------------------- ')
# Next use FLDA Classifier
classify4 = LinearDiscriminantAnalysis()
# Train the model
classify4.fit(x_train, y_train)
# Use the model on the test data
predicted4 = classify4.predict(x_test)
print("LD", predicted4)
ld = metrics.accuracy_score(y_test, predicted4) * 100
print("The accuracy score using the FLDA is ->")
print(metrics.accuracy_score(y_test, predicted4))
print('---------------------------------------------- ')
# Next use SVM
classify5 = svm.LinearSVC()
# Train the model
classify5.fit(x_train, y_train)
# Use the model on the test data
predicted5 = classify5.predict(x_test)
svmdt = metrics.accuracy_score(y_test, predicted5) * 100
print("The accuracy score using the svm is ->")
print(metrics.accuracy_score(y_test, predicted5))
print('---------------------------------------------- ')
list = []
list.clear()
list.append(int(nb))
list.append(int(ld))
list.append(int(svmdt))
view(list)
def performancealg():
datainput = pd.read_csv("trainingset.csv")
# Set the outcome and dedlete it
y = datainput['Placed']
del datainput['Placed']
# Split data into Test & Training set where test data is 30% & training data is 70%
x_train, x_test, y_train, y_test = train_test_split(datainput,
y,
test_size=0.3)
# MLPClassifier() Classifier
classify3 = MLPClassifier()
# Train the model
classify3.fit(x_train, y_train)
# Use the model on the test data
predicted3 = classify3.predict(x_test)
lr = metrics.accuracy_score(y_test, predicted3) * 100
print("The accuracy score using the ANN Classifier is ->")
print(metrics.accuracy_score(y_test, predicted3))
print('---------------------------------------------- ')
# DecisionTreeClassifier() Classifier
classify4 = DecisionTreeClassifier()
# Train the model
classify4.fit(x_train, y_train)
# Use the model on the test data
predicted4 = classify4.predict(x_test)
dt = metrics.accuracy_score(y_test, predicted4) * 100
print("The accuracy score using DecisionTreeClassifier() is ->")
print(metrics.accuracy_score(y_test, predicted4))
print('---------------------------------------------- ')
# SVM()
classify5 = SVC(gamma='auto')
# Train the model
classify5.fit(x_train, y_train)
# Use the model on the test data
predicted5 = classify5.predict(x_test)
rf = metrics.accuracy_score(y_test, predicted5) * 100
print("The accuracy score using the SVM() is ->")
print(metrics.accuracy_score(y_test, predicted5))
print('---------------------------------------------- ')
list = []
list.clear()
list.append(lr)
list.append(dt)
list.append(rf)
view(list)
def graphdef(self):
view()