ott.core.bar_problems.BarycenterProblem

class ott.core.bar_problems.BarycenterProblem(y=None, b=None, weights=None, cost_fn=None, epsilon=None, debiased=False, segment_ids=None, num_segments=None, indices_are_sorted=None, num_per_segment=None, max_measure_size=None)

Definition of a linear regularized OT problem and some tools.

Parameters

• y (Optional[ndarray]) – a matrix merging the points of all measures.

• b (Optional[ndarray]) – a vector containing the weights (within each masure) of all the points

• weights (Optional[ndarray]) – weights of the barycenter problem (size num_segments)

• cost_fn (Optional[CostFn]) – cost function used.

• epsilon (Optional[ndarray]) – epsilon regularization used to solve reg-OT problems.

• debiased (bool) – whether the problem is debiased, in the sense that the regularized transportation cost of barycenter to itself will be considered when computing gradient. Note that if the debiased option is used, the barycenter size (used in call function) needs to be smaller than the max_measure_size parameter below, for parallelization to operate efficiently.

• segment_ids (Optional[ndarray]) – describe for each point to which measure it belongs.

• num_segments (Optional[ndarray]) – total number of measures

• indices_are_sorted (Optional[bool]) – flag indicating indices in segment_ids are sorted.

• num_per_segment (Optional[ndarray]) – number of points in each segment, if contiguous.

• max_measure_size (Optional[int]) – max number of points in each segment (for efficient jit)

Methods

add_slice_for_debiased(y, b)

rtype Tuple[Optional[ndarray], Optional[ndarray]]

Attributes

flattened_b	rtype Optional[ndarray]
flattened_y	rtype Optional[ndarray]
max_measure_size	rtype int
num_segments	rtype int
segmented_y_b	rtype Tuple[Optional[ndarray], Optional[ndarray]]
weights	rtype ndarray