def sklearnNaiveBayes(test, train, structFile):
model=GaussianNB()
le=preprocessing.LabelEncoder()
target_train=train['class']
inputs_train=train.drop('class',axis='columns')
target_test=test['class']
inputs_test=test.drop('class',axis='columns')
inputs_train=fit_transforms(inputs_train)
model.fit(inputs_train,target_train)
# save model to file
filename = 'NaiveBayesSKlearn_model.sav'
joblib.dump(model, filename)
inputs_test=fit_transforms(inputs_test)
print("sklearnNaiveBayes accuracy:",model.score(inputs_test,target_test),"%")
def TestTrainFitPlot(train, test):
# Setup arrays to store train and test accuracies
# Split into training and test set
target_train = train['class']
inputs_train = train.drop('class', axis='columns')
target_test = test['class']
inputs_test = test.drop('class', axis='columns')
inputs_train = fit_transforms(inputs_train)
inputs_test = fit_transforms(inputs_test)
knn = KNeighborsClassifier()
knn.fit(inputs_train, target_train)
# save model to file
filename = 'KNN_model.sav'
joblib.dump(knn, open(filename, 'wb'))
# Check Accuracy Score
print('KNN Accuracy: {}'.format(
round(knn.score(inputs_test, target_test), 3)))
# Enum Loop, accuracy results using range on 'n' values for KNN Classifier
'''
def ID3SKlearn_algorithm(test,train,structFile):
train=fit_transforms(train)
train_target= train['class']
train_feature=train.drop('class', axis='columns')
test=fit_transforms(test)
test_target= test['class']
test_feature=test.drop('class', axis='columns')
tree=DecisionTreeClassifier(criterion='entropy',max_depth=100).fit(train_feature,train_target)
# save model to file
filename = 'ID3SKlearn_model.sav'
joblib.dump(tree, filename)
prediction = tree.predict(test_feature)
print("ID3SKlearn_algorithm accuracy is: ",tree.score(test_feature,test_target)*100,"%")
def KNNClassifier(test, train, structFile):
encode = fit_transforms(train)
encode_ = fit_transforms(test)
x = feature("class", encode)
y = feature("class", encode_)
TestTrainFitPlot(train, test)
