deepfold.losses.procrustes

Functions

flatten_batch_dims(tensor, end_dim)
kabsch(x, y[, weights, compute_scaling])	Returns the rigid transformation and the optimal scaling that best align an input list of points x to a target list of points y, by minimizing the sum of square distance.
procrustes(m[, force_rotation, ...])	Returns the orthonormal matrix minimizing Frobenius norm.
rigid_points_registration(x, y[, weights, ...])	Returns the rigid transformation and the optimal scaling that best align an input list of points x to a target list of points y, by minimizing the sum of square distance.
rigid_vectors_registration(x, y[, weights, ...])
speical_procrustes(m[, regularization, ...])
svd(m)	Singular value decomposition.
unflatten_batch_dims(tensor, batch_shape)

Classes

Procrustes(*args, **kwargs)