1

我用 Keras 创建了一个自定义初始化程序。部分代码是:

def my_init(shape):
    P = tf.get_variable("P", shape=shape,    initializer = tf.contrib.layers.xavier_initializer())
    return P

model = Sequential()
model.add(Conv2D(32, kernel_size=(5, 5),strides=(1, 1), padding='same', input_shape = input_shape, kernel_initializer = my_init))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, kernel_size=(1, 1) , strides=(1, 1) , padding='same' , kernel_initializer = my_init))

当在卷积层中第二次调用“my_init”初始化程序时,它会抛出此错误:

Variable P already exists, disallowed. Did you mean to set reuse=True in VarScope? Originally defined at:

不允许重用变量 P。有没有办法在每次调用中创建一个新变量?

4

1 回答 1

1

您可以尝试使用 Keras 中可用的 Xavier 初始化程序,名称为glorot_uniformglorot_normal.

在这里看到它们:https ://keras.io/initializers/

model.add(Conv2D(32, kernel_size=(1, 1) , strides=(1, 1) , 
          padding='same' , kernel_initializer =keras.initializers.glorot_uniform())
于 2017-09-27T15:59:07.670 回答