def build_model(self):
conv_base = MobileNet(
include_top=False,
weights='imagenet',
input_shape=config.input_shape
)
conv_base.trainable = False
print(conv_base.summary())
inputs = Input(shape=config.input_shape)
x = conv_base(inputs)
x = Flatten()(x)
x = Dense(units=128, activation=mobilenet['activation'])(x)
x = Dropout(rate=0.5)(x)
outputs = Dense(units=config.num_classes, activation=mobilenet['outact'])(x)
self.model = Model(inputs=inputs, outputs=outputs)
input_shape=(32, 32, 3))
# print(model.weights)
# ============== 전처리 ===================
x_train = preprocess_input(x_train)
x_test = preprocess_input(x_test)
x_train = x_train.astype('float32') / 255. # 전처리
x_test = x_test.astype('float32') / 255. # 전처리
#다중분류 y원핫코딩
y_train = to_categorical(y_train) #(50000, 10)
y_test = to_categorical(y_test) #(10000, 10)
# ============== 모델링 =====================
MobileNetV2.trainable = False # 훈련을 안시키겠다, 저장된 가중치 사용
MobileNetV2.summary()
# 즉, 16개의 레이어지만 연산되는 것은 13개 이고 그래서 len=26개
print(len(MobileNetV2.weights)) # 26
print(len(MobileNetV2.trainable_weights)) # 0
model = Sequential()
model.add(MobileNetV2) # 3차원 -> layer 26개
model.add(Flatten())
model.add(Dense(10))
model.add(Dense(5))
model.add(Dense(10, activation='softmax'))
model.summary()
#3. 컴파일, 훈련
model.compile(loss='categorical_crossentropy',
optimizer='adam',
