compute_polygon_jacobian

ClosedCurve.compute_polygon_jacobian(params: ndarray) → ndarray

Compute the Jacobian of the discretized polygon vertices.

Uses automatic differentiation (autograd) to compute: J[i,j] = d(polygon[i]) / d(params[j])

where polygon is flattened to shape (M*2,) with x and y coordinates interleaved

Parameters:

paramsnp.ndarray

Optimization parameters (shape: (n_params,)).

Returns:

np.ndarray

Jacobian matrix of shape (M*2, n_params) where M is the number of polygon points.