Я использую алгоритм нечеткого искусства для применения к моим данным, и все работает нормально в случае результатов, но когда дело доходит до отображения результата, интерпретатор говорит: RuntimeError: matplotlib не поддерживает генераторы в качестве входных данных.
Кстати, я использую python в качестве языка программирования и matplotlib для отображения результатов.
import numpy as np
from sys import argv
import matplotlib.pyplot as plt
from random import shuffle
def scaledList(list):
min = None
max = None
for i in range(len(list)):
if min is None or min > list[i]:
min = list[i]
if max is None or max < list[i]:
max = list[i]
for i in range(len(list)):
list[i] = (float)(list[i] - min)/(max - min)
#print "min = " + str(min) + ", max = " + str(max)
return list
def complementCodedConversion(list1, list2):
list = []
for i in range(len(list1)):
complementVector = (list1[i], list2[i], 1-list1[i], 1-list2[i])
list.append(complementVector)
return list
def fuzzyAnd(tuple1, tuple2):
ls = []
for i in range(len(tuple1)):
minimum = min(tuple1[i], tuple2[i])
ls.append(minimum)
#print(ls)
return ls
def printData(list):
for i in range(len(list)):
print (list[i])
def myPlot(dataList, boxList):
x = map(lambda item:item[0], dataList)
y = map(lambda item:item[1], dataList)
plt.scatter(x, y)
for i in range(len(boxList)):
(a,b,c,d) = boxList[i]
plt.plot([a,c,c,a,a], [b, b, d, d, b])
plt.show()
training = np.loadtxt(fname="C:\\Users\\Ali\Desktop\\fuzzy-art-neural-network-master\\TrainingData.txt")
testing = np.loadtxt(fname="C:\\Users\\Ali\Desktop\\fuzzy-art-neural-network-master\\TestingData.txt")
trainingData = complementCodedConversion(scaledList(training[:, 2].tolist()), scaledList(training[:, 3].tolist()))
testingData = complementCodedConversion(scaledList(testing[:, 2].tolist()), scaledList(testing[:, 3].tolist()))
#printData(trainingData)
#parameters setting
#learning rate
beta = 1
#vigilance
rho = 0.8
alpha = 0.000001
categoryList = []
#shuffle(trainingData)
while True:
#shuffle(trainingData)
len1 = len(categoryList)
for i in range(len(categoryList)):
tjList = []
for j in range(len(categoryList)):
summation1 = fuzzyAnd(trainingData[i], categoryList[j])
#summation2 = summation1[i] / (alpha+sum(categoryList[j]))
ss1 = sum(categoryList[j])
ss2 = alpha + ss1
summation2 = summation1[i] / ss2
tjList.append((summation2))
tjList = sorted(tjList, key=lambda item:item[1])
noMatchFlag = True
while len(tjList) != 0 :
(index, value) = tjList.pop(0)
if sum(fuzzyAnd(trainingData[i], categoryList[index]))/sum(trainingData[i]) >= rho :
categoryList[index] = map(lambda x, y: x*beta + y*(1-beta),
fuzzyAnd(trainingData[i], categoryList[index]), categoryList[index])
noMatchFlag = False
break
if noMatchFlag:
categoryList.append(trainingData[i])
novellist = []
for i in range(len(testingData)):
tjList = []
for j in range(len(categoryList)):
tjList.append((j, sum(fuzzyAnd(testingData[i], categoryList[j]))/(alpha+sum(categoryList[j]))))
tjList = sorted(tjList, key=lambda item:item[1])
noMatchFlag = True
while len(tjList) != 0 :
(index, value) = tjList.pop(0)
if sum(fuzzyAnd(testingData[i], categoryList[index]))/sum(testingData[i]) >= rho :
noMatchFlag = False
break
if noMatchFlag:
novellist.append(i)
print ("***************")
printData(categoryList)
print ("***************")
print ("Novel list:")
print (novellist)
myPlot(trainingData, categoryList)
#print "novelist: ", len(novellist)
len2 = len(categoryList)
if len2 == len1 :
break
#print "number of centers: ", len2
Результат интерпретатора на изображении ниже:
Как я могу решить эту проблему, чтобы построить свои результаты,Заранее спасибо