tf.sub(x, y, name=None) Returns x - y element-wise. NOTE: Sub supports broadcasting

tf.contrib.distributions.Distribution.variance(name='variance') Variance.

tf.contrib.ffmpeg.decode_audio(contents, file_format=None, samples_per_second=None, channel_count=None) Create an op that decodes

tf.contrib.util.constant_value(tensor) Returns the constant value of the given tensor, if efficiently calculable.

tf.nn.rnn_cell.MultiRNNCell.state_size

tf.contrib.distributions.MultivariateNormalCholesky.mean(name='mean') Mean.

tf.contrib.distributions.MultivariateNormalDiag.parameters Dictionary of parameters used by this Distribution.

tf.contrib.distributions.Laplace.std(name='std') Standard deviation.

tf.contrib.distributions.MultivariateNormalCholesky.log_prob(value, name='log_prob') Log probability density/mass function (depending

tf.contrib.distributions.WishartCholesky.scale_operator_pd Wishart distribution scale matrix as an OperatorPD.