keras：Sequential()模型

---

• 简单的层的堆叠，本层只从上一层接收，只传递到下一层。
• 单输入，单输出

model = keras.Sequential([
        layers.Input(shape=(4,)),
        layers.Dense(2, activation="relu"),
        layers.Dense(3, activation="relu"),
        layers.Dense(4),
])

model.summary()
'''
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #
=================================================================
dense (Dense)                (None, 2)                 10
_________________________________________________________________
dense_1 (Dense)              (None, 3)                 9
_________________________________________________________________
dense_2 (Dense)              (None, 4)                 16
=================================================================
Total params: 35
Trainable params: 35
Non-trainable params: 0
_________________________________________________________________
'''

不在模型中指定输入层keras.Input(shape=(4,)),，算是还没建完的半成品，model.summary()就会报错。