fuzzc

A simple python module to Use fuzzy logic. All the methods return the Fuzzy_Set class. Which is easy to use compared to traditional numpy arrays.

Hint that the module is still in proggress. Any contribution is welcome. The idea is to provide a utility, with the ability to compete with matlab.

Available Functions Up to now.

add, substract, multiplication, comparative operatiors ( >, <, etc), show, support, crossover, is_fuzzy_singleton, center, plot_membership, height, alpha_cut, core, is_normal, is_convex, yager and sugeno complements.

To get started check: https://github.com/KiLJ4EdeN/fuzzc/blob/master/examples/fuzzc.ipynb

Prerequisites:

1 - Python

2 - Numpy

3 - Matplotlib

To see an example usage:

1 - Install Using pip or optionally clone the repo and build from source.

pip install fuzzc

or

git clone https://github.com/KiLJ4EdeN/fuzzc/
cd fuzzc
python3 setup.py build

2 - Import the module and try some functionalities:

import fuzzc.core as fuzzc
import numpy as np

# create a set of points in the set.
fuzzy_set = np.linspace(0, 100, 1, dtype=np.int32)
print(fuzzy_set)
# create a random membership function.
membership_fn = np.around(np.random.rand(len(fuzzy_set)), decimals=1)
# create the Fuzzy_Set instance.
myset = fuzzc.Fuzzy_Set(fuzzy_set=fuzzy_set, membership_fn=membership_fn)
# Use the utilities.
myset.show()
myset.is_fuzzy_singleton()

3 - Operators Defined for fuzzy sets.

import fuzzc.core as fuzzc
import numpy as np

fuzzy_set1 = np.array([130, 140, 150, 160, 170, 180, 190, 200, 210])
membership_fn1 = np.array([0., 0.1, 0.25, 0.5, 0.9, 0.5, 0.25, 0.1, 0])
fuzzy1 = Fuzzy_Set(fuzzy_set=fuzzy_set1, membership_fn=membership_fn1)
fuzzy_set2 = np.array([130, 140, 150, 160, 170, 180, 190, 200, 210])
membership_fn2 = np.array([1., 0.9, 0.3, 0.6, 1.0, 0.6, 0.5, 0.3, 0.4])
fuzzy2 = Fuzzy_Set(fuzzy_set=fuzzy_set2, membership_fn=membership_fn2)
print(fuzzy1 + fuzzy2)
print(fuzzy1 > fuzzy2)
print(fuzzy1 < fuzzy2)
print(fuzzy1 - fuzzy2)
print(fuzzy1 * fuzzy2)
print(fuzzy1 == fuzzy2)
print(fuzzy1 != fuzzy2)

4 - A more generilized example.

import fuzzc.core as fuzzc
import numpy as np

# height fuzzy example with plot.
# define height ranges
fuzzy_set = np.array([130, 140, 150, 160, 170, 180, 190, 200, 210])
# create a custom membership function for average height (avg = 170).
membership_fn = np.array([0., 0.1, 0.25, 0.5, 1.0, 0.5, 0.25, 0.1, 0])
myset = fuzzc.Fuzzy_Set(fuzzy_set=fuzzy_set, membership_fn=membership_fn)
# show the members and their respective membership.
myset.show()
# plot the membership function.
myset.plot_membership()
# find the height.
myset.height()
# extract the cross over points and the support.
supp_set = myset.support()
supp_set.show()
cross_set = myset.crossover()
cross_set.show()
# find the center of the set.
myset.center()
# Alpha cuts are also available.
alpha_cut = myset.alpha_cut(0.5)
alpha_cut.show()
# The core of a fuzzy set 
# is where the memberships are maximized( ==1 ).
core = myset.core()
core.show()
# Also notice that the set and the membership function can be assesed anytime.
print(myset.fuzzy_set)
print(myset.membership_fn)

5 - complement methods.

import fuzzc.core as fuzzc
import numpy as np

fuzzy_set = np.array([130, 140, 150, 160, 170, 180, 190, 200, 210])
# create a custom membership function for average height (avg = 170)
membership_fn = np.array([0., 0.1, 0.25, 0.5, 1.0, 0.5, 0.25, 0.1, 0])
fuzzy = Fuzzy_Set(fuzzy_set=fuzzy_set, membership_fn=membership_fn)
(fuzzy.complement(method=None)).show()
(fuzzy.complement(method='sugeno', l=3)).show()
(fuzzy.complement(method='yager', w=3)).show()