def _get_backbone_model(self):
backbone_model = None
if self.backbone_name == 'resnet18':
backbone_model = ResNet18(input_shape=self.input_shape, include_top=False, weights='imagenet')
if self.backbone_name == 'resnet34':
backbone_model = ResNet34(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'mobilenetv2':
backbone_model = MobileNetV2(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'efficientnetb0':
backbone_model = EfficientNetB0(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'efficientnetb1':
backbone_model = EfficientNetB1(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'efficientnetb3':
backbone_model = EfficientNetB3(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'efficientnetb5':
backbone_model = EfficientNetB5(input_shape=self.input_shape, include_top=False, weights='imagenet')
elif self.backbone_name == 'densenet':
backbone_model = DenseNet121(input_shape=self.input_shape, include_top=False, weights='imagenet')
return backbone_model
def __init__(self, model_size , imag_size, num_cls):
try:
if model_size == "B0": model = EfficientNetB0(weights = 'imagenet', input_shape = (imag_size,imag_size,3), include_top = False)
elif model_size == "B3": model = EfficientNetB3(weights = 'imagenet', input_shape = (imag_size,imag_size,3), include_top = False)
elif model_size == "B5": model = EfficientNetB5(weights = 'imagenet', input_shape = (imag_size,imag_size,3), include_top = False)
elif model_size == "B7": model = EfficientNetB7(weights = 'imagenet', input_shape = (imag_size,imag_size,3), include_top = False)
ENet_out = model.output
ENet_out = Flatten()(ENet_out)
Hidden1_in = Dense(1024, activation="relu")(ENet_out)
Hidden1_in = Dropout(0.5)(Hidden1_in)
predictions = Dense(units = num_cls, activation="softmax")(Hidden1_in)
self.model_f = Model(input = model.input, output = predictions)
self.model_f.compile(optimizers.Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=1e-08), loss='categorical_crossentropy', metrics=[metrics.mae, metrics.categorical_accuracy])
except:
print("only B0/B3/B5/B7 allowed")
def get_model(self,
class_names,
model_name="EfficientNetB3",
use_base_weights=True,
weights_path=None,
input_shape=None):
if use_base_weights is True:
base_weights = "imagenet"
else:
base_weights = None
if input_shape is None:
input_shape = self.models_[model_name]["input_shape"]
img_input = Input(shape=input_shape)
base_model = EfficientNetB3(include_top=False,
input_tensor=img_input,
input_shape=input_shape,
weights=base_weights,
pooling="avg")
x = base_model.output
predictions = Dense(len(class_names),
activation="sigmoid",
name="predictions")(x)
model = Model(inputs=img_input, outputs=predictions)
if weights_path == "":
weights_path = None
if weights_path is not None:
print(f"load model weights_path: {weights_path}")
model.load_weights(weights_path)
return model
target_size=(img_height,
img_width),
batch_size=n_batch_size,
shuffle=True,
subset='training')
valid_generator = train_datagen.flow_from_directory('./train/data',
target_size=(img_height,
img_width),
batch_size=n_batch_size,
shuffle=True,
subset='validation')
# 以訓練好的 EfficientNetB3 為基礎來建立模型
net = EfficientNetB3(input_shape=(img_height, img_width, 3),
include_top=False,
weights='imagenet',
pooling='max')
# 增加 Dense layer 與 softmax 產生個類別的機率值
x = net.output
# x = Dense(1024, activation='relu')(x)
# x = Dropout(0.5)(x)
# x = Dense(128, activation='relu')(x)
output_layer = Dense(3, activation='softmax', name='softmax')(x)
# 設定要進行訓練的網路層
model = Model(inputs=net.input, outputs=output_layer)
print('\n')
print('Trainable layers:')
for x in model.trainable_weights:
def trainModel(self):
print(os.listdir(self.__data_directory))
class_nb = len(os.listdir(self.__data_directory))
train_generator = self.__image_generator.flow_from_directory(
directory = self.__data_directory,
target_size=(self.__image_size,self.__image_size),
subset="training",
batch_size=self.__batch_size,
shuffle=True,
class_mode= "categorical",
seed=42
)
val_generator = self.__image_generator.flow_from_directory(
directory = self.__data_directory,
target_size=(self.__image_size,self.__image_size),
subset="validation",
batch_size=self.__batch_size,
shuffle=True,
class_mode= "categorical",
seed=42
)
if self.__model_type== "EfficientNetB3":
pretrainedModel = EfficientNetB3(
weights='imagenet',
input_shape=(self.__image_size, self.__image_size, 3),
include_top=False,
pooling='max'
)
else :
raise Exception('{} model is not a member of our available pretrained models'.format(self.__model_type))
model = Sequential()
model.add(pretrainedModel)
model.add(Dense(units = 120, activation='relu'))
model.add(Dense(units = 120, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(units = class_nb, activation='softmax'))
model.summary()
model.compile(
optimizer=Adam(lr=0.0001, clipnorm=1, clipvalue=5),
loss='categorical_crossentropy',
metrics=['accuracy']
)
logdir = "logs/" + datetime.now().strftime("%Y%m%d-%H%M%S")+ "_" + self.__model_name
callbacks = [
#EarlyStopping(monitor='val_loss', patience= 10),
ModelCheckpoint(filepath= self.__model_name + '_best_model.h5', monitor='val_loss', save_best_only=True),
TensorBoard(log_dir=logdir)]
history = model.fit_generator(
train_generator,
epochs = self.__epoch_number,
#steps_per_epoch = 20,
validation_data = val_generator,
#validation_steps = 5,
callbacks = callbacks
)
acc = history.history['accuracy']
val_acc = history.history['val_accuracy']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(1,len(acc) + 1)
fig=plt.figure(figsize=(15, 5))
fig.add_subplot(1, 2, 1)
plt.plot(epochs,acc,'g',label = 'Training Accuracy')
plt.plot(epochs,val_acc,'r',label = 'Validation Accuracy')
plt.title('Training and Validation Accuracy')
plt.legend()
fig.add_subplot(1, 2, 2)
plt.plot(epochs,loss,'g',label = 'Training loss')
plt.plot(epochs,val_loss,'r',label = 'Validation Loss')
plt.title('Training and Validation Loss')
plt.legend()
plt.show()
# serialize weights to HDF5
model.save(self.__model_name +'.h5')
print("Saved final model : " + self.__model_name +'.h5')
print("Saved best model considering val_loss : " + self.__model_name + '_best_model.h5')
def test_efficientnet(self):
model = EfficientNetB3(weights=None)
vis = KerasArchitectureVisualizer()
vis.visualize(model)
batch_size=128,
shuffle=True,
class_mode="binary")
val_generator = datagen.flow_from_dataframe(dataframe=df,
directory=data_dir,
x_col="filename",
y_col="category",
target_size=(32, 32),
subset="validation",
batch_size=64,
shuffle=True,
class_mode="binary")
efficient_net = EfficientNetB3(weights='imagenet',
input_shape=(32, 32, 3),
include_top=False,
pooling='max')
model = Sequential()
model.add(efficient_net)
model.add(Dense(units=120, activation='relu'))
model.add(Dense(units=120, activation='relu'))
model.add(Dense(units=1, activation='sigmoid'))
model.summary()
model.compile(optimizer=Adam(lr=0.0001),
loss='binary_crossentropy',
metrics=['accuracy'])
filename = './checkpoints/efficient-weights.h5'
log_dir = "./logs/fit/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S")