def test_cross_validate_without_loading_data():
k_fold = 5
mlf.select_classifier("Decision Tree")
ret_val = mlf.cross_validate(k_fold)
assert isinstance(ret_val, str)
def test_invalid_input_select_classifier():
invalid_input = 37246
print(mlf.CLASSIFIER)
mlf.select_classifier(invalid_input)
print(mlf.CLASSIFIER)
assert mlf.CLASSIFIER is None
def test_load_classifier():
mlf.select_classifier("Decision Tree")
classifier_before_saving = mlf.CLASSIFIER
mlf.save_classifier()
mlf.load_classifier()
assert classifier_before_saving == mlf.CLASSIFIER
def test_save_classifier():
mlf.select_classifier("Decision Tree")
existing_classifiers = os.listdir("Classifiers")
pickle_path = os.path.split(mlf.save_classifier())[1]
# Check if a file was created and if is is unique
assert (os.path.isfile(pickle_path)
and (pickle_path not in existing_classifiers))
def test_cross_validate_with_invalid_k_fold():
k_fold = "invalid!"
malware_rate = 0.33
mlf.load_json_data(malware_rate)
mlf.select_classifier("Decision Tree")
ret_val = mlf.cross_validate(k_fold)
assert isinstance(ret_val, str)
def test_cross_validate_with_classifier():
k_fold = 5
malware_rate = 0.33
mlf.load_json_data(malware_rate) # TODO: value error
mlf.select_classifier("Decision Tree")
ret_val = mlf.cross_validate(k_fold)
assert isinstance(ret_val, numpy.ndarray)
def test_train_classifier_with_classifier():
malware_rate = 0.45
partition_size = 0.92
mlf.load_json_data(malware_rate)
mlf.partition_data(partition_size)
mlf.select_classifier("Decision Tree")
mlf.train_classifier()
def test_valid_input_select_classifier():
valid_input = "Decision Tree"
ret_val = mlf.select_classifier(valid_input)
assert (isinstance(ret_val, str) and mlf.CLASSIFIER is not None)
def test_calculate_classifier_accuracy_with_classifier():
malware_rate = 0.33
partition_size = 0.33
mlf.load_json_data(malware_rate)
mlf.partition_data(partition_size)
mlf.select_classifier("Decision Tree")
mlf.train_classifier()
ret_val = mlf.calculate_classifier_accuracy()
assert isinstance(ret_val, float)
def test_classify_new_data_with_classifier():
malware_rate = 0.33
partition_size = 0.33
mlf.load_json_data(malware_rate)
mlf.partition_data(partition_size)
mlf.select_classifier("Decision Tree")
mlf.train_classifier()
ret_val = mlf.classify_new_data()
assert isinstance(ret_val, numpy.ndarray)
def test_display_confusion_matrix_pass():
malware_rate = 0.33
partition_size = 0.33
mlf.REPORT.create_new_report_directory()
mlf.load_json_data(malware_rate)
mlf.partition_data(partition_size)
mlf.select_classifier("Decision Tree")
mlf.train_classifier()
mlf.classify_new_data()
mlf.display_confusion_matrix()
def classifier_selection_menu():
global CLASSIFIER_NAME
if CLASSIFIER_NAME:
classifier_functionalities_menu()
else:
click.echo("\nClassifier Selection Menu\n")
user_opt = prompt_user_with_menu(mlf.CLASSIFIER_DICT, "Select a classifier")
if user_opt:
CLASSIFIER_NAME = mlf.select_classifier(user_opt)
classifier_functionalities_menu()