3dec543fe5e96ffd27705ab3e31e288d106903e3,tensorlayer/layers.py,BatchNormLayer,__init__,#BatchNormLayer#Any#Any#Any#Any#Any#Any#Any#Any#,1783

Before Change

                self.outputs = act( tf.nn.batch_normalization(self.inputs, mean, variance, beta, gamma, epsilon) )

            // variables = tf.get_collection(TF_GRAPHKEYS_VARIABLES, scope=vs.name)  // 8 params in TF12 if zero_debias=True
            variables = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope=vs.name)    // 2 params beta, gamma
                // variables = [beta, gamma, moving_mean, moving_variance]

            // print(len(variables))

After Change

            // variables = tf.get_collection(TF_GRAPHKEYS_VARIABLES, scope=vs.name)  // 8 params in TF12 if zero_debias=True
            // variables = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope=vs.name)    // 2 params beta, gamma
                // variables = [beta, gamma, moving_mean, moving_variance]
            variables = [beta, gamma]

            // print(len(variables))
            // for idx, v in enumerate(variables):
            //     print("  var {:3}: {:15}   {}".format(idx, str(v.get_shape()), v))

Instances