tf.contrib.framework.get_variables_by_name(given_name, scope=None) Gets the list of variables that were given that name.

tf.contrib.framework.get_graph_from_inputs(op_input_list, graph=None) Returns the appropriate graph to use for the given inputs

tf.contrib.framework.get_model_variables(scope=None, suffix=None) Gets the list of model variables, filtered by scope and/or suffix

tf.contrib.framework.model_variable(*args, **kwargs) Gets an existing model variable with these parameters or creates a new one

tf.is_non_decreasing(x, name=None) Returns True if x is non-decreasing. Elements

class tf.contrib.framework.VariableDeviceChooser Device chooser for variables. When using a parameter

tf.contrib.framework.assign_from_values(var_names_to_values) Creates an assignment operation from a given mapping.

tf.contrib.framework.get_or_create_global_step(graph=None) Returns and create (if necessary) the global step variable.

tf.contrib.framework.deprecated(date, instructions) Decorator for marking functions or methods deprecated. This

tf.contrib.framework.assign_from_checkpoint(model_path, var_list) Creates an operation to assign specific variables from a checkpoint