# Module: ed.util

Defined in edward/util/__init__.py.

## Classes

class Progbar

## Functions

check_data(...): Check that the data dictionary passed during inference and

check_latent_vars(...): Check that the latent variable dictionary passed during inference and

compute_multinomial_mode(...): Compute the mode of a Multinomial random variable.

copy(...): Build a new node in the TensorFlow graph from org_instance,

dot(...): Compute dot product between a 2-D tensor and a 1-D tensor.

get_ancestors(...): Get ancestor random variables of input.

get_blanket(...): Get Markov blanket of input, which consists of its parents, its

get_children(...): Get child random variables of input.

get_control_variate_coef(...): Returns scalar used by control variates method for variance reduction in

get_descendants(...): Get descendant random variables of input.

get_parents(...): Get parent random variables of input.

get_session(...): Get the globally defined TensorFlow session.

get_siblings(...): Get sibling random variables of input.

get_variables(...): Get parent TensorFlow variables of input.

is_independent(...): Assess whether a is independent of b given the random variables in

random_variables(...): Return all random variables in the TensorFlow graph.

rbf(...): Radial basis function kernel, also known as the squared

set_seed(...): Set seed for both NumPy and TensorFlow.

to_simplex(...): Transform real vector of length (K-1) to a simplex of dimension K

transform(...): Transform a continuous random variable to the unconstrained space.

with_binary_averaging(...): Inspired by scikit-learn’s _average_binary_score function: