def __init__(self, dir_z, dir_x, nb_channels=3):
assert get_nb_files(dir_z) == get_nb_files(dir_x) #如果条件不满足，返回错误
assert nb_channels in [1, 3]
self.nb_files = get_nb_files(dir_z)
self.nb_channels = nb_channels
self.dir_z = dir_z
self.dir_x = dir_x
def __init__(self, dir_z, dir_x, nb_channels=3):
assert get_nb_files(dir_z) == get_nb_files(dir_x)
assert nb_channels in [1, 3]
self.nb_files = get_nb_files(dir_z)
self.nb_channels = nb_channels
self.dir_z = dir_z
self.dir_x = dir_x
def __init__(self, dir_z, dir_x, nb_channels=3):
assert get_nb_files(dir_z) == get_nb_files(
dir_x), "data infomation: {}".format(dir_z)
assert nb_channels in [1, 3]
self.firstchannel = net_config['first_channels']
self.nb_files = get_nb_files(dir_z)
self.nb_channels = nb_channels
self.dir_z = dir_z
self.dir_x = dir_x
self.filename = os.listdir(dir_x)
def __init__(self, dir_z, dir_x, nb_channels=3):
# check the number of files in X(means original image) and Z(the implicit vector after vector) are equal?
assert get_nb_files(dir_z) == get_nb_files(dir_x), \
"The files numbers are not equal. \n Please check infomation in {0} and {1}\n".format(dir_z, dir_x)
assert nb_channels in [1, 3]
self.nb_files = get_nb_files(dir_z)
self.nb_channels = nb_channels
self.dir_z = dir_z
self.dir_x = dir_x
self.filename = os.listdir(dir_x)
def __init__(self, dir_x, nb_channels=3):
assert nb_channels in [1, 3]
self.nb_channels = nb_channels
self.nb_files = get_nb_files(dir_x)
self.dir_x = dir_x
def TransferLearning(args):
"""
Performs training.
"""
train_dir = args.train_folder
val_dir = args.validation_folder
nb_train_samples = utils.get_nb_files(args.train_folder)
nb_val_samples = utils.get_nb_files(args.validation_folder)
nb_classes = utils.get_labels(args.train_folder, args.validation_folder)
nb_epochs = int(args.nb_epoch)
batch_size = int(args.batch_size)
base_architecture = args.base_architecture
model_load = args.model_load
# Define base layer
if base_architecture == 'VGG16':
base_model = applications.VGG16(weights='imagenet', include_top=False)
layers_to_freeze = 10
elif base_architecture == 'VGG19':
base_model = applications.VGG19(weights='imagenet', include_top=False)
layers_to_freeze = 11
elif base_architecture == 'InceptionV3':
base_model = applications.InceptionV3(weights='imagenet', include_top=False)
layers_to_freeze = 172 # TODO: understand how many levels!
elif base_architecture == 'ResNet50':
base_model = applications.ResNet50(weights='imagenet', include_top=False)
elif base_architecture == 'Xception':
base_model = applications.Xception(weights='imagenet', include_top=False)
model = replace_classification_layer(base_model, nb_classes, 1024)
train_datagen = ImageDataGenerator(
rescale = 1./255,
fill_mode='nearest')
test_datagen = ImageDataGenerator(
rescale = 1./255,
fill_mode='nearest')
train_generator = train_datagen.flow_from_directory(
train_dir,
target_size=(img_height, img_width),
batch_size=batch_size,
class_mode='categorical')
validation_generator = test_datagen.flow_from_directory(
val_dir,
target_size=(img_height, img_width),
batch_size=batch_size,
class_mode='categorical')
transfer_learning(base_model, model, model_load)
history_tl = model.fit_generator(
train_generator,
nb_epoch=nb_epochs,
samples_per_epoch=nb_train_samples,
validation_data=validation_generator,
nb_val_samples=nb_val_samples,
class_weight='auto',
callbacks=args.callbacks)
utils.plot_training(history_tl)
setup_fine_tuning(model, layers_to_freeze, model_load)
history_ft = model.fit_generator(
train_generator,
samples_per_epoch=nb_train_samples,
nb_epoch=nb_epochs,
validation_data=validation_generator,
nb_val_samples=nb_val_samples,
class_weight='auto',
callbacks=args.callbacks)
# NOTE
model.save(os.path.join(os.getcwd(), 'models', args.output_model_file))
utils.plot_training(history_ft)