Workaround for bug in TF 1.14

      # Create learnable weights for the generators and discriminators.

      gen_alpha = layers.Variable(np.ones((1, n_generators)), dtype=tf.float32)

      gen_weights = Softmax()(gen_alpha([]))

      # We pass an input to the Variable layer to work around a bug in TF 1.14.

      gen_weights = Softmax()(gen_alpha([self.noise_input]))

      discrim_alpha = layers.Variable(

          np.ones((1, n_discriminators)), dtype=tf.float32)

      discrim_weights = Softmax()(discrim_alpha([]))

      discrim_weights = Softmax()(discrim_alpha([self.noise_input]))

      # Compute the weighted errors