如何在训练时获得层权重？

如何解决如何在训练时获得层权重？

我有一个模型，我想在定义自定义损失函数时获取特定层的权重矩阵以使用它。

有没有办法在模型内部获得特定层的权重？

附言我目前正在使用 tensorflow 2 和 keras 功能 API。我测试了 How do I get the weights of a layer in Keras? 方法，但没有奏效。

P.P.S.通过使用上述方法，我收到以下错误：

AttributeError                            Traceback (most recent call last)
<ipython-input-26-e0bd481102a7> in <module>
      1 A_DENSE = Dense(1,use_bias = True,name = "A_DENSE")(INPUT)
----> 2 A_DENSE.get_weights()

AttributeError: 'Tensor' object has no attribute 'get_weights'

P.P.P.S.如下所述，结合自定义回调和 get_weights 解决了这个问题。祝和我有类似情况的人好运。

解决方法

您可以编写自定义 Callback 并在每次 epoch 结束时使用它。我展示了它用于打印权重，但您可以将其用作自定义损失的一部分。

class CustomCallback(keras.callbacks.Callback):
    def on_epoch_end(self,epoch,logs=None):
        rand_int = tf.random.uniform((),2,dtype=tf.int32)
        print(rand_int)
        
model.fit(X,y epochs = 10,batch_size = 20,validation_split=0.1,callbacks=[CustomCallback()])

更多详情here。

---

例如，这是一个伪代码，用于在每个 epoch 之后打印 weights and biases 的 layer[1]。您可以按照自己喜欢的方式设置该功能。

from tensorflow.keras import layers,Model,callbacks

class CustomCallback(callbacks.Callback):
    def on_epoch_end(self,logs=None):
        print(' ')
        print(' ')
        print(model.layers[1].get_weights())
        

X,y = np.random.random((10,5)),np.random.random((10,))

inp = layers.Input((5,))
x = layers.Dense(3)(inp)
out = layers.Dense(1)(x)

model = Model(inp,out)

model.compile(loss='MAE',metrics=['accuracy'])
model.fit(X,y,callbacks=[CustomCallback()],epochs=3)

Epoch 1/3
1/1 [==============================] - ETA: 0s - loss: 0.2346 - accuracy: 0.0000e+00 
 
[array([[ 0.16518219,-0.44628695,-0.07702655],[-0.1993848,0.03855793,-0.62964785],[ 0.5592851,-0.28281152,-0.23358124],[ 0.05242977,0.4023881,-0.19522922],[ 0.07936202,-0.40436065,0.10003945]],dtype=float32),array([ 0.01530731,-0.01565045,-0.01581042],dtype=float32)]
1/1 [==============================] - 0s 2ms/step - loss: 0.2346 - accuracy: 0.0000e+00
Epoch 2/3
1/1 [==============================] - ETA: 0s - loss: 0.2337 - accuracy: 0.0000e+00 
 
[array([[ 0.16814367,-0.4492649,-0.08000461],[-0.19710523,0.03622784,-0.6319782 ],[ 0.55797213,-0.28144714,-0.23221655],[ 0.05509637,0.3996864,-0.19793113],[ 0.07731982,-0.40226308,0.10213734]],array([ 0.01846951,-0.01881272,-0.01897269],dtype=float32)]
1/1 [==============================] - 0s 7ms/step - loss: 0.2337 - accuracy: 0.0000e+00
Epoch 3/3
1/1 [==============================] - ETA: 0s - loss: 0.2322 - accuracy: 0.0000e+00 
 
[array([[ 0.16706704,-0.448164,-0.07889817],[-0.19894598,0.0381193,-0.63007975],[ 0.5558067,-0.27921563,-0.22997847],[ 0.05663134,0.3981127,-0.19951159],[ 0.07536169,-0.400249,0.10415838]],dtype=float32)]
1/1 [==============================] - 0s 2ms/step - loss: 0.2322 - accuracy: 0.0000e+00