def save_model_to_file(filename,
architecture,
weights='imagenet',
input_shape=(299, 299, 3)):
base_path = os.getcwd()
model_path = os.path.join(base_path, filename)
# Create and save network file
if not os.path.exists(model_path):
# Initialise Keras object
if architecture == 'inceptionv3':
mymodel = applications.InceptionV3(
weights=weights, input_tensor=Input(shape=input_shape))
elif architecture == 'resnet152v2':
mymodel = applications.ResNet152V2(
weights=weights, input_tensor=Input(shape=input_shape))
elif architecture == 'inceptionresnetv2':
mymodel = applications.InceptionResNetV2(
weights=weights, input_tensor=Input(shape=input_shape))
else:
print('Invalid architecture')
return
mymodel.save(os.path.join(base_path, filename))
def create_model(nclass):
base_model = applications.ResNet152V2(weights='imagenet',
include_top=False,
input_shape=input_shape)
base_model.trainable = False
model = Sequential()
model.add(base_model)
model.add(Flatten())
model.add(BatchNormalization())
model.add(Dense(2048, activation='relu'))
model.add(BatchNormalization())
model.add(Dense(1024, activation='relu'))
model.add(BatchNormalization())
model.add(Dense(nclass, activation='softmax'))
return model
def create_resnet_model():
base = applications.ResNet152V2(include_top=False,
weights=None,
input_shape=(112, 112, 3),
pooling='max')
x = base.output
x = Dropout(0.4)(x)
x = Dense(2048, activation='relu')(x)
x = Dense(2048, activation='relu')(x)
x = Dense(1024, activation='relu')(x)
x = Dense(1024, activation='relu')(x)
x = Dense(512, activation='relu')(x)
x = Dense(512, activation='relu')(x)
x = Dense(256, activation='relu')(x)
x = Dense(256, activation='relu')(x)
output = Dense(1, activation='linear')(x)
model = Model(inputs=base.input, outputs=output)
return model
x_train, y_train = load_data(os.getcwd() + "/archive/seg_train/")
x_test, y_test = load_data(os.getcwd() + "/archive/seg_test/")
x_train, x_valid, y_train, y_valid = train_test_split(x_train,
y_train,
test_size=0.15,
random_state=0)
# %--------------------------------------------- Training ----------------------------------------------------------
# Resnet
BATCH_SIZE = 512
EPOCHS = 10
LR = 1e-3
res = applications.ResNet152V2(input_shape=(150, 150, 3),
weights='imagenet',
include_top=False)
res.trainable = False
print('Resnet pre trained model is loaded ....')
model = Sequential([
res,
Flatten(),
Dense(400, activation='tanh'),
Dropout(0.5),
BatchNormalization(),
Dense(6, activation='softmax')
])
early_stopping_callback = keras.callbacks.EarlyStopping(monitor='val_loss',
patience=3)