pyGrC

A set of methods about granular computing which is realized by python 3

基于Python 3 实现的一系列粒计算方法

methods

fuzzy c-means (模糊c均值聚类)

The code is in fcm.py

[U, V] = fcmAO(data, c, m, threshold)

data: list of lists where every list is a point
c: the number of clusters
m: the parameter to control fuzziness
threshold: the stop condition
U: the membership of every point belonging to every center
V: list of lists where every list is a center

代码见fcm.py

[U, V] = fcmAO(data, c, m, threshold)

data: list的list，其中每个list是一个点
c: 聚类个数
m: 控制模糊程度的参数
threshold: 迭代停止条件
U: 每个点隶属于每个类别的隶属度矩阵
V: list的list，其中每个list是一个类别中心

k-means (k均值聚类)

The code is in kmeans.py

[U, V] = kmeans(data, k, threshold)

data: list of lists where every list is a point
k: the number of clusters
threshold: the stop condition
U: the membership of every point belonging to every center which is 0 or 1
V: list of lists where every list is a center

代码见kmeans.py

[U, V] = kmeans(data, k, threshold)

data: list的list，其中每个list是一个点
k: 聚类个数
threshold: 迭代停止条件
U: 每个点隶属于每个类别的隶属度矩阵，属于为1，不属于为0
V: list的list，其中每个list是一个类别中心

k-medians (k中位数聚类)

The code is in kmedian.py

[U, V] = kmedians(data, k, threshold)

data: list of lists where every list is a point
k: the number of clusters
threshold: the stop condition
U: the membership of every point belonging to every center which is 0 or 1
V: list of lists where every list is a center

代码见kmedian.py

[U, V] = kmedians(data, k, threshold)

data: list的list，其中每个list是一个点
k: 聚类个数
threshold: 迭代停止条件
U: 每个点隶属于每个类别的隶属度矩阵，属于为1，不属于为0
V: list的list，其中每个list是一个类别中心

principle of justifiable granularity (合理粒度原则)

The code is in pojg.py

granule = pojgMedE(data, alpha)

data: list of numbers which are used to form information granule
alpha: the parameter to adjust the importance of specificity
granule: the information granule which is [lower bound, median, upper bound]

代码见pojg.py

granule = pojgMedE(data, alpha)

data: 需要被粒化的数字list
alpha: 控制specificity的参数
granule: 信息粒，形式为 [下界, 中位数, 上界]

rough set (粗糙集近似)

The code is in rs.py

Roughset = rs(classes, Set)

classes: list of sets or list of lists, where every sets or list is a class
Set: the targer set which needs to be approximated
Roughset: list of two sets, which are lower approximation and upper approximation

代码见rs.py

Roughset = rs(classes, Set)

classes: set的list或list的list, 其中每个set或list是一个类
Set: 需要被近似的集合
Roughset: 下近似与上近似组成的list

decision theoretic rough set (决策粗糙集近似)

I am coding 正在开发中

1075161149 / pyGrC