CODEWITHSUNDEEP
tensorflowmachine-learningkeraslayer
NewToCoding
NewToCoding

Reputation: 303

how to create trainable weight variable for custom keras layer

I am implementing custom average pooling layer, where each neuron computes the mean of its inputs, then multiplies the result by a learnable coefficient and adds a learnable bias term, then finally applies the activation function

from tensorflow.keras.layers import Layer

from keras import backend as K


class Average_Pooling_Layer(Layer):
    def __init__(self, output_dimension, **kwargs):
        super(Average_Pooling_Layer, self).__init__(**kwargs)
        self.output_dimension = output_dimension

    def build(self, input_shape):

        self.weights = self.add_weight(name='weights2',
                              shape=(input_shape[0], 
                               int(self.output_dimension[0]), 
                               int(self.output_dimension[1]), 
                               int(self.output_dimension[2])),
                               initializer='uniform',
                               trainable=True)
    super(Average_Pooling_Layer, self).build(input_shape)

   def call(self, inputs):
        return K.tanh((inputs * self.weights))

    def compute_output_shape(self, input_shape):
        return (input_shape)

Code Usage

model = tf.keras.Sequential()
stride = 1
c1 = model.add(Conv2D(6, kernel_size=[5,5], strides=(stride,stride), padding="valid", input_shape=(32,32,1), 
                  activation = 'tanh'))
s2_before_activation = model.add(AveragePooling2D(pool_size=(2, 2), strides=(2, 2)))
s2 = model.add(Average_Pooling_Layer(output_dimension = (14, 14, 6)))

I am getting error as "Failed to convert object of type to Tensor. Contents: (Dimension(None), 14, 14, 6). Consider casting elements to a supported type." "None" is batch size, which I am getting from previous layer.

How to solve this?

Upvotes: 2

Views: 3901

Answers (1)

giser_yugang
giser_yugang

Reputation: 6176

Your error is caused by the data type. input_shape[0] returns <class 'tensorflow.python.framework.tensor_shape.Dimension'> instead of int.

You can replace input_shape[0] with tf.TensorShape(input_shape).as_list()[0]. But your data dimension is not right and you have to adjust and modify it according to your needs.

Edit

If you get error as "can't set attribute", you should rename your weight variable instead of self.weights. For example, change to self.weights_new.

Upvotes: 1

Related Questions