def align_faces():
start = time.time()
align_mtcnn('your_dataset', 'face_align')
end = time.time()
#total_minute = (end - start) / 60
#print('Aligning excecution time: ' + str(end - start) + ' second')
#print('Total minutes:', total_minute)
messagebox.showinfo(
"Information", "The alignment is complete \n" +
"Aligning excecution time: " + str(format(end - start, ".2f")) + "s")
model.fit(emb_array, labels)
# Create a list of class names
class_names = [cls.name.replace('_', ' ') for cls in dataset]
# Saving classifier model
with open(classifier_filename_exp, 'wb') as outfile:
pickle.dump((model, class_names), outfile)
print('Saved classifier model to file "%s"' %
classifier_filename_exp)
def split_dataset(dataset, min_nrof_images_per_class,
nrof_train_images_per_class):
train_set = []
test_set = []
for cls in dataset:
paths = cls.image_paths
# Remove classes with less than min_nrof_images_per_class
if len(paths) >= min_nrof_images_per_class:
np.random.shuffle(paths)
train_set.append(
ImageClass(cls.name, paths[:nrof_train_images_per_class]))
test_set.append(
ImageClass(cls.name, paths[nrof_train_images_per_class:]))
return train_set, test_set
if __name__ == '__main__':
align_mtcnn('your_face', 'face_align')
train('face_align/', 'models/20180402-114759.pb', 'models/your_model.pkl')
# Create a list of class names
class_names = [cls.name.replace('_', ' ') for cls in dataset]
# Saving classifier model
with open(classifier_filename_exp, 'wb') as outfile:
pickle.dump((model, class_names), outfile)
print('Saved classifier model to file "%s"' %
classifier_filename_exp)
def split_dataset(dataset, min_nrof_images_per_class,
nrof_train_images_per_class):
train_set = []
test_set = []
for cls in dataset:
paths = cls.image_paths
# Remove classes with less than min_nrof_images_per_class
if len(paths) >= min_nrof_images_per_class:
np.random.shuffle(paths)
train_set.append(
ImageClass(cls.name, paths[:nrof_train_images_per_class]))
test_set.append(
ImageClass(cls.name, paths[nrof_train_images_per_class:]))
return train_set, test_set
if __name__ == '__main__':
align_mtcnn('input_face', 'output_face')
train('output_face/', 'models/20180402-114759.pb', 'models/your_model.pkl')
# Create a list of class names
class_names = [cls.name.replace('_', ' ') for cls in dataset]
# Saving classifier model
with open(classifier_filename_exp, 'wb') as outfile:
pickle.dump((model, class_names), outfile)
print('Saved classifier model to file "%s"' %
classifier_filename_exp)
def split_dataset(dataset, min_nrof_images_per_class,
nrof_train_images_per_class):
train_set = []
test_set = []
for cls in dataset:
paths = cls.image_paths
# Remove classes with less than min_nrof_images_per_class
if len(paths) >= min_nrof_images_per_class:
np.random.shuffle(paths)
train_set.append(
ImageClass(cls.name, paths[:nrof_train_images_per_class]))
test_set.append(
ImageClass(cls.name, paths[nrof_train_images_per_class:]))
return train_set, test_set
if __name__ == '__main__':
align_mtcnn('data_face', 'face_align')
train('face_align/', 'models/20180402-114759.pb', 'models/your_model.pkl')
def TrainImages():
align_mtcnn('Datasets', 'face_align')
train_data('face_align/', 'models/20180402-114759.pb',
'models/your_model.pkl')
res = "Image Trained"
message.configure(text=res)
model.fit(emb_array, labels)
# Create a list of class names
class_names = [cls.name.replace('_', ' ') for cls in dataset]
# Saving classifier model
with open(classifier_filename_exp, 'wb') as outfile:
pickle.dump((model, class_names), outfile)
print('Saved classifier model to file "%s"' %
classifier_filename_exp)
def split_dataset(dataset, min_nrof_images_per_class,
nrof_train_images_per_class):
train_set = []
test_set = []
for cls in dataset:
paths = cls.image_paths
# Remove classes with less than min_nrof_images_per_class
if len(paths) >= min_nrof_images_per_class:
np.random.shuffle(paths)
train_set.append(
ImageClass(cls.name, paths[:nrof_train_images_per_class]))
test_set.append(
ImageClass(cls.name, paths[nrof_train_images_per_class:]))
return train_set, test_set
if __name__ == '__main__':
align_mtcnn('Datasets', 'face_align')
train('face_align/', 'models/20180402-114759.pb', 'models/your_model.pkl')
