if __name__ == '__main__':
args = get_args()
# Augmentation Data
data_gen_args = dict(rotation_range=0.2,
width_shift_range=0.05,
height_shift_range=0.05,
shear_range=0.05,
zoom_range=0.05,
horizontal_flip=True,
fill_mode='nearest')
myGene = trainGenerator(args.batch_size,
args.data_dir,
'image',
'label',
data_gen_args,
save_to_dir=None)
if args.use_pretrained:
# Pretrain model
model = keras.models.load_model('models/hairnet_matting.hdf5')
else:
model = Hairnet.get_model()
model.compile(optimizer=Adam(lr=args.lr),
loss='binary_crossentropy',
metrics=['accuracy'])
model_checkpoint = ModelCheckpoint('models/hairnet_matting.hdf5',
monitor='loss',
from keras.callbacks import ModelCheckpoint
import keras
from data.load_data import trainGenerator
from nets import Hairnet
if __name__ == '__main__':
BATCH_SIZE = 4
DATA_PATH = 'data/'
data_gen_args = dict(rotation_range=0.2,
width_shift_range=0.05,
height_shift_range=0.05,
shear_range=0.05,
zoom_range=0.05,
horizontal_flip=True,
fill_mode='nearest')
myGene = trainGenerator(BATCH_SIZE, DATA_PATH, 'image', 'label', data_gen_args,
save_to_dir=None)
model = Hairnet.get_model()
# Pretrain model
# model = keras.models.load_model('hairnet_matting3.hdf5')
model.compile(optimizer=Adam(lr=1e-4), loss='binary_crossentropy', metrics=['accuracy'])
model_checkpoint = ModelCheckpoint('hairnet_matting.hdf5', monitor='loss', verbose=1, save_best_only=True)
model.fit_generator(myGene, callbacks=[model_checkpoint], steps_per_epoch=2000, epochs=30)