network.py 文件源码

def batch_norm(x, phase_train):
    """
    Batch normalization on convolutional maps.
    Args:
        x:           Tensor, 4D BHWD input maps
        n_out:       integer, depth of input maps
        phase_train: boolean tf.Variable, true indicates training phase
        scope:       string, variable scope
        affn:      whether to affn-transform outputs
    Return:
        normed:      batch-normalized maps
    Ref: http://stackoverflow.com/questions/33949786/how-could-i-use-batch-normalization-in-tensorflow/33950177
    """
    name = 'batch_norm'
    with tf.variable_scope(name):
        phase_train = tf.convert_to_tensor(phase_train, dtype=tf.bool)
        n_out = int(x.get_shape()[3])
        beta = tf.Variable(tf.constant(0.0, shape=[n_out], dtype=x.dtype),
                           name=name+'/beta', trainable=True, dtype=x.dtype)
        gamma = tf.Variable(tf.constant(1.0, shape=[n_out], dtype=x.dtype),
                            name=name+'/gamma', trainable=True, dtype=x.dtype)

        batch_mean, batch_var = tf.nn.moments(x, [0,1,2], name='moments')
        ema = tf.train.ExponentialMovingAverage(decay=0.9)
        def mean_var_with_update():
            ema_apply_op = ema.apply([batch_mean, batch_var])
            with tf.control_dependencies([ema_apply_op]):
                return tf.identity(batch_mean), tf.identity(batch_var)
        mean, var = control_flow_ops.cond(phase_train,
                                          mean_var_with_update,
                                          lambda: (ema.average(batch_mean), ema.average(batch_var)))
        normed = tf.nn.batch_normalization(x, mean, var, beta, gamma, 1e-3)
    return normed