fuzzycorr 0.1.1

A numpy implementation of Robust Rank Correlation Coefficients.

fuzzycorr

A numpy implementation of Robust Rank Correlation Coefficients (fuzzy correlation) based on paper:

Bodenhofer, U., and F. Klawonn. "Robust rank correlation coefficients on the basis of fuzzy."
Mathware & Soft Computing 15.1 (2008): 5-20.

This implementation is experimental and need future optimization and testing.

Installation

This package will be avaliable soon on pip.

Basic usage

from fuzzycorr import fuzzy_correlation_factory
from fuzzycorr.strict_orderings import lukasiewicz_strict_ordering_factory
from fuzzycorr.t_norms import godel

# create strict fuzzy ordering or supply own one
strict_ordering = lukasiewicz_strict_ordering_factory(r=0.2)

# create fuzzy correlation function with tnorm
fuzzy_corr = fuzzy_correlation_factory(strict_ordering, godel)

# load data
x = np.random.random(10)
y = np.random.random(10)

# calculate fuzzy correlation
fuzzy_corr(x, y)

Visit example Jupiter Notebook.