multidim.fast_algorithms

This Cython module contains core algorithms for multidimensional data. In particular, some distance caching and the cover-tree friends matching.

If a function requires its own explicit iterator in a loop, probably it should go here. Anything that is just a vectorized operation in NumPy or Pandas can go in plain python elsewhere.

See http://docs.cython.org/src/tutorial/np.html for ideas on writing efficient code in Cython.

This module is compiled by either of these commands

• python setup.py install (as called by pip for standard installation and use)
• python setup.py build_ext --inplace (as run by developers for code testing)

Notes

Automatically-generated online documentation will never see the C-only functions that are defined with cdef in Cython, as shown in http://cython.readthedocs.io/en/latest/src/reference/extension_types.html Because of the limitations of Sphinx, you’ll have to simply view the sourcecode for further information. For this module, these C-only functions include

• is_partition_bool()
• is_partition_list()

Copyright

• This file is part of https://github.com/geomdata/gda-public/
• 2015, 2016, 2017 by Geometric Data Analytics, Inc. (http://geomdata.com)
• AGPL license. See LICENSE or https://github.com/geomdata/gda-public/blob/master/LICENSE

Functions

cdist(XA, XB[, metric, p, V, VI, w])	Computes distance between each pair of the two collections of inputs.
check_heights	Verify that cells of dimension dim have weight greater than their boundaries of dimension dim - 1.
covertree_adopt_or_liberate	adopt an orphan using type-1 friends.
covertree_befriend321	Set friends3 using inheritance filter
covertree_exchange_teens	For a given center of a CoverTree, determine which of its children are
distance_cache_None	Get bulk distances, using no storage cache.
distance_cache_dict	Get bulk distances, using dictionary storage cache.
distance_cache_dok	Get bulk distances, using dictionary storage cache.
distance_cache_numpy	Get bulk distances, using NumPy storage cache.
edges_from_dists
entropy	Compute the base-N entropy of normalized bins [bin0, bin1, …, bin(N-1)]
euclidean(u, v)	Computes the Euclidean distance between two 1-D arrays.
gaussian	Generate value for a multivariate Gaussian PDF.
gaussian_fit	Fit a normalized Gaussian to this cloud using singular value decomposition.
is_partition_bool	Check that bigblock is partitioned by list_of_small_blocks
is_partition_list	You have a big list and a list of small lists.
label_means	Given a list of points, a list of labels, and a list of weights, compute the weighted mean for each label.
label_weights	Weights-per-label of a subset with labels and weights.