def t_predict(features, labels, cameras):
log_file = '/driving_log.csv'
log_path = './data'
skiprows = 1
# load the csv log file
print("Camera: ", cameras)
print("Log path: ", log_path)
print("Log file: ", log_file)
column_names = [
'center', 'left', 'right', 'steering', 'throttle', 'brake', 'speed'
]
data_df = pd.read_csv(log_path + '/' + log_file,
names=column_names,
skiprows=skiprows)
data_count = len(data_df)
# initialise data extract
t_features = []
t_labels = []
row = data_df.iloc[np.random.randint(data_count - 1)]
steering = getattr(row, 'steering')
image = load_image(log_path, getattr(row, cameras[0]))
t_features.append(image)
t_labels.append(steering)
clf = GaussianNB()
clf.fit(features, labels)
pred = clf.pred(t_features)
print(accuracy_score(pred, t_labels))
### labels_train and labels_test are the corresponding item labels
features_train, features_test, labels_train, labels_test = preprocess()
#########################################################
### your code goes here ###
# Defining Classifier - Gaussian Naive Bayes
clf = GaussianNB()
# Fitting the training data set
# training the test data set labels
t0 = time()
clf.fit(features_train,lables_train)
print("training naive bayes:", round(time()-t0, 3), "s")
#predicting the test dataset labels using the trained statistics
t0 = time()
pred = clf.pred(features_test)
print("predicting naive bayes:", round(time()-t0, 3), "s")
accuracy = clf.score(features_test,labels_test)
print(accuracy)
#########################################################
