我有一个简单的神经网络
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, MaxPool2D
model = Sequential([
Conv2D(16,(3,3),padding='same', input_shape=(1,28,28),data_format='channels_first'),
MaxPooling2D((3,3), data_format='channels_first')
])
print(model.summary())
模型总结是
Model: "sequential_4"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d_2 (Conv2D) (None, 16, 28, 28) 160
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 16, 9, 9) 0
=================================================================
Total params: 160
Trainable params: 160
Non-trainable params: 0
_________________________________________________________________
None
并编译为
opt = tf.keras.optimizers.Adam(learning_rate=0.005)
model.compile(optimizer=opt,
loss=tf.keras.losses.BinaryCrossentropy(),
metrics=[tf.keras.metrics.BinaryAccuracy(),
tf.keras.metrics.MeanAbsoluteError()]
)
现在,当我打印模型属性时,它给出了一个关于指标的空列表。为什么会这样?
print(f"{model.metrics}\n{model.optimizer},\n,{model.loss}\n{model.optimizer.lr}")
这是输出
[]
<tensorflow.python.keras.optimizer_v2.adam.Adam object at 0x000001FB10F2EDC8>,
,<tensorflow.python.keras.losses.BinaryCrossentropy object at 0x000001FB10F2EEC8>
<tf.Variable 'learning_rate:0' shape=() dtype=float32, numpy=0.005>
最佳答案
目前,this是 Tensorflow 2.2.0 版中的一个错误。以后可能会修复
关于python - Tensorflow Keras 指标未显示,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/62116136/