如何在keras中使用可学习的权重进行矩阵乘法？

Question

我有一个像下面这样的模型。我想在最后添加一个可学习权重矩阵，它被初始化为我传递给函数 create_model.

的变量 matrix

为了直观地了解我想做什么，假设矩阵应该是我传递给模型的那个，但我觉得它可以还是在训练的时候进行微调。因此，我希望它被初始化为我传递的值，然后在训练过程中进行细化。

下面的代码有效，但正如您从 model.summary() 输出中看到的那样，矩阵乘法不包含可学习的权重，这让我认为权重矩阵未微调。

我做错了什么？

 def create_model(num_columns, matrix):
    inp_layer = tfl.Input((num_columns,))
    dense = tfl.Dense(512, activation = 'relu')(inp_layer)
    dense = tfl.Dense(256, activation = 'relu')(dense)
    dense = tfl.Dense(128, activation = 'relu')(dense)

    va = tf.Variable(matrix, dtype = tf.float32)
    dense = K.dot(dense, va )
    
    model = tf.keras.Model(inputs = inp_layer, outputs = dense)
    model.compile(optimizer='adam', loss=['binary_crossentropy'])
    model.summary()
    return model

matrix = np.random.randint(0,2,(128, 206)) # In reality, this is not random, but it has sensed values
num_columns = 750
model = create_model(num_columns,matrix)

Answer 1

你可以简单地使用一个没有偏差的密集层来做这个乘法。建立模型后，我用您提供的矩阵更改了感兴趣的权重

def create_model(num_columns, matrix):
    
    inp_layer = Input((num_columns,))
    x = Dense(512, activation = 'relu')(inp_layer)
    x = Dense(256, activation = 'relu')(x)
    x = Dense(128, activation = 'relu')(x)
    dense = Dense(206, use_bias=False)(x)
    
    model = Model(inputs = inp_layer, outputs = dense)
    model.compile(optimizer='adam', loss=['binary_crossentropy'])
    
    model.set_weights(model.get_weights()[:-1] + [matrix])
    model.summary()
    
    return model

matrix = np.random.randint(0,2,(128, 206)) # In reality, this is not random, but it has sensed values
num_columns = 750
model = create_model(num_columns,matrix)

检查

(model.get_weights()[-1] == matrix).all() # True

这样权重就可以fine-tuned

如何在keras中使用可学习的权重进行矩阵乘法？

How to do matrix multiplication with learnable weights in keras?

python

matrix-multiplication

keras

tensorflow