grid_distance_graph

tangles.util.graph.similarity.grid_distance_graph(grid_shape: tuple[int, int], max_dist: float = 3.0, normalize: bool = False, dist: str = 'euklid') → csr_array

A function to create a grid-like graph where every vertex is connected to all neighbors that are close (in grid coordinates).

Returns a 2-dimensional matrix, where the index of each axis represents the index of a square. The value is the distance between the squares in the grid.

Parameters

grid_shapeshape

A shape. Given as a tuple containing exactly two integers.

max_distfloat

Maximal distance of two connected vertices.

normalizebool

If True, distances are normalized.

dist{‘euklid’, ‘manhattan’}

The metric that will be used to calculate the distances.

Returns

sparse.csr_array

A similarity graph between the pixels.