当我在Keras中运行model.summary()
时,我得到以下输出:
Model: "sequential"_________________________________________________________________Layer (type) Output Shape Param # =================================================================output (Dense) (1, 1) 1 =================================================================Total params: 1Trainable params: 1Non-trainable params: 0_________________________________________________________________None
底部的None是什么意思?
我运行了以下代码:
input_numpy_array=np.array([1])model = keras.models.Sequential()input_layer = keras.layers.Input(shape=input_numpy_array.shape, name='input', batch_size=1)model.add(input_layer)output_layer = keras.layers.Dense(1, use_bias=False, name='output', batch_size=1)model.add(output_layer)print(model.summary())
回答:
model.summary()
不需要print
语句:它会自动打印,并且返回None
。当你打印model.summary()
时,你实际上是在打印None
。
x = model.summary()x == None
True