This module contains Python bindings to a C-implementation of "An efficiently computable metric for comparing polygonal shapes" (Arkin et al., 1991).

pip install turning_function

Call turning_function.distance(shape_a, shape_b, brute_force_updates=False) where shape_a/shape_b are Nx2-shaped lists of points (not numpy arrays). This function will return four values: distance, theta, ht_err, slope_err. The error terms will be zero if brute_force_updates is False. Note that there is a maximum number of points (turning_function.max_points) a shape can hold.

import turning_function
import random
def random_shape(num_points: int):
	return [(random.random(), random.random()) for _ in range(num_points)]
shape_a = random_shape(turning_function.max_points)
shape_b = random_shape(42)
distance, theta, ht_err, slope_err = turning_function.distance(shape_a, shape_b, brute_force_updates=False)
print('Distance: ', distance)

The original source has this disclaimer:

Implementation (c) Eugene K. Ressler 91, 92 This source may be freely distributed and used for non-commercial purposes, so long as this comment is attached to any code copied or derived from it.