7

So I'm trying to implement my own layer in Keras, using the provided example:

class MyLayer(Layer):

    def __init__(self, output_dim, **kwargs):
        self.output_dim = output_dim
        super(MyLayer, self).__init__(**kwargs)

    def build(self, input_shape):
        # Create a trainable weight variable for this layer.
        self.kernel = self.add_weight(name='kernel', 
                                      shape=(input_shape[1], self.output_dim),
                                      initializer='uniform',
                                      trainable=True)
        super(MyLayer, self).build(input_shape)  # Be sure to call this somewhere!

    def call(self, x):
        return K.dot(x, self.kernel)

    def compute_output_shape(self, input_shape):
        return (input_shape[0], self.output_dim)

I noticed that what's in call gets called at build time when I do:

model.add(MyLayer(4, input_dim=(1))

This happens without calling fit, train, predict, etc...

Why?

gotch4
  • 13,093
  • 29
  • 107
  • 170

1 Answers1

5

It will be called at the time you added layer to model to check if the shape is valid or not.

If you want to see where it is called, change your code:

import sys
import traceback

class MyLayer(Layer):
    ....
    def call(self, x):
         traceback.print_stack(file=sys.stdout)
         return K.dot(x, self.kernel)
An Phú
  • 457
  • 3
  • 10