Crate classif

Library for one-dimensional data classification and simple statistics, especially methods used for cartographic purposes such as “quantiles breaks” or “Jenks natural breaks”.

This library allows to computed break values according to a few methods (listed in the Classification Enum). Theses break values can be computed by creating a new BoundsInfo struct.

If the input values are already sorted and you don’t care about error checking you can directly use the get_{jenks|quantiles|etc...}_breaks functions.

This library also provide a few basic statistical functionnalities, allowing to compute mean value, kurtosis value, standard deviation, variance, root of mean square, etc.

Modules

stats

Basic statistical functionnalities: mean, standard deviation, kurtosis, variance, etc.

Structs

BoundsInfo

A struct containing the bounds computed at its creation and some basic statistical informations : minimum, maximum and mean value.

Enums

Classification

The various type of classification methods availables.

Functions

get_arithmetic_breaks

Compute the “arithmetic progression” breaks on a list of sorted values.

get_equal_interval

Compute the equal interval breaks on a list of sorted values.

get_head_tail_breaks

Compute the “Head-Tail” breaks on a list of sorted values (to be used on heavily right skewed distributions).

get_jenks_breaks

Compute the “Natural Breaks” on a list of sorted values, based on Jenks optimization.

get_quantiles

Compute the quantiles breaks on a list of sorted values.

get_tail_head_breaks

Compute the “Tail-Head” breaks on a list of sorted values (its actually just the inverse of the Head-Tail method, to be used on heavily left skewed distributions).