-
Notifications
You must be signed in to change notification settings - Fork 0
/
EXAFSgen - 10path - diff breed - diff add children - diff parentpool - rdm fitness.py
326 lines (261 loc) · 9.7 KB
/
EXAFSgen - 10path - diff breed - diff add children - diff parentpool - rdm fitness.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
import larch
import random
from larch_plugins.xafs import feffdat
from larch import Interpreter
import operator
import numpy as np
from operator import itemgetter
from larch_plugins.io import read_ascii
from larch_plugins.xafs import autobk
import datetime
import time
import matplotlib as mpl
mpl.use('Agg')
import matplotlib.pyplot as plt
mylarch = Interpreter()
#range for rdm num generator
rangeA = (np.linspace(5,150,146) * 0.01).tolist()
rangeB = (np.linspace(-500,500,1001) * 0.01).tolist()
rangeC = (np.linspace(-20,20,41) * 0.01).tolist()
rangeD = (np.linspace(1,15,15) * 0.001).tolist()
rangeA.append(0)
#rangeB.append(0)
#rangeC.append(0)
#rangeD.append(0)
front = 'Cu Data/path Data/feff'
end = '.dat'
intervalK = (np.linspace(80,340,261)).tolist()
def fitness(indi,exp):
loss = 0
yTotal = [0]*(261)
for i in range(1,10):
if i < 10:
filename = front+'000'+str(i)+end
elif i< 100:
filename = front+'00'+str(i)+end
else:
filename = front+'0'+str(i)+end
path=feffdat.feffpath(filename, s02=str(indi[i-1][0]), e0=str(indi[i-1][1]), sigma2=str(indi[i-1][2]), deltar=str(indi[i-1][3]), _larch=mylarch)
feffdat._path2chi(path, _larch=mylarch)
y = path.chi
for k in intervalK:
yTotal[int(k)-80] += y[int(k)-80]
global g
for j in range(len(yTotal)):
loss = loss + (yTotal[j]*g.k[j+80]**2 - exp[j]*g.k[j+80]**2)**2
return loss
def generateACombo():
a = random.choice(rangeA)
global b
c = random.choice(rangeC)
d = random.choice(rangeD)
return [a,b,c,d]
def generateIndi():
indi = []
for i in range(10):
indi.append(generateACombo())
return indi
def generateFirstGen(genSize):
gen = []
i = 0
while i < genSize:
gen.append(generateIndi())
i+=1
return gen
def computePerfPop(pop,exp):
populationPerf = {}
for individual in pop:
individualTuple = tuple(tuple(x) for x in individual)
populationPerf[individualTuple] = fitness(individual, exp)
return sorted(populationPerf.items(), key = operator.itemgetter(1), reverse=False)
def selectFromPopulation(populationSorted, best_sample, lucky_few):
nextBreeders = []
# fitSum = 0
# current = 0
# for indi in populationSorted:
# fitSum += indi[1]
# pick = random.uniform(0, fitSum)
# for indi in populationSorted:
# current += indi[1]
# if current < pick:
# newIndi = []
# for combo in indi[0]:
# newIndi.append(list(combo))
# nextBreeders.append(newIndi)
for i in range(best_sample):
nextBreeders.append(populationSorted[i])
for i in range(lucky_few):
j = random.randint(best_sample,len(populationSorted)-1)
nextBreeders.append(populationSorted[j])
random.shuffle(nextBreeders)
return nextBreeders
def createChild(individual1, individual2):
child = []
global diffCounter
global chance_of_mutation
if diffCounter > 10:
print("******************Different Breeding******************")
chance_of_mutation = 40
diffCounter = 0
for i in range(len(individual1)):
if (int(100 * random.random()) < 50):
child.append(individual1[i][0:2] + individual2[i][2:4])
else:
child.append(individual2[i][0:2] + individual1[i][2:4])
else:
chance_of_mutation = 20
for i in range(len(individual1)):
j = random.randint(0,1)
if j == 0:
child.append(individual1[i])
elif j == 1:
child.append(individual2[i])
return child
def createChildren(breeders, number_of_child):
nextPopulation = []
for i in range(int(len(breeders)/2)):
for j in range(number_of_child):
# b1 = random.randint(0,len(breeders)-1)
# b2 = random.randint(0,len(breeders)-1)
# nextPopulation.append(createChild(breeders[b1],breeders[b2]))
nextPopulation.append(createChild(breeders[i], breeders[len(breeders) -1 -i]))
return nextPopulation
def mutateIndi(indi):
indi = generateIndi()
return indi
def mutatePopulation(population, chance_of_mutation, chance_of_mutation_e0):
mutateTime = 0
for i in range(len(population)):
if random.random() * 100 < chance_of_mutation:
mutateTime+=1
population[i] = mutateIndi(population[i])
if random.random() * 100 < chance_of_mutation_e0:
e0 = random.choice(rangeB)
print("Mutate e0 to:", e0)
for i in range(len(population)):
for j in population[i]:
j[1] = e0
print("Mutate Times:", mutateTime)
return population
def nextGeneration (firstGeneration, exp, best_sample, lucky_few, number_of_child, chance_of_mutation, chance_of_mutation_e0):
st = datetime.datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S')
global genNum
global historyBest
global bestDiff
global diffCounter
genNum+=1
print(st)
print("Gen:", genNum)
populationTupleSorted = computePerfPop(firstGeneration, exp)
populationSorted = []
print("Best Fit:", populationTupleSorted[0][1])
for indi in populationTupleSorted:
newIndi = []
for combo in indi[0]:
#print("--",combo)
newIndi.append(list(combo))
populationSorted.append(newIndi)
bestDiff = abs(populationTupleSorted[0][1]-historyBest)
historyBest = populationTupleSorted[0][1]
if bestDiff < 0.1:
diffCounter += 1
else:
diffCounter = 0
print("2nd Fit:",populationTupleSorted[1][1])
print("3rd Fit:",populationTupleSorted[2][1])
print("4th Fit:",populationTupleSorted[3][1])
print("Last Fit:",populationTupleSorted[len(populationTupleSorted)-1][1])
print("Different from last best fit:",bestDiff)
print("Best fit combination:\n",populationTupleSorted[0][0])
# if genNum%1 == 0:
# print("Best fit combination:\n",populationTupleSorted[0][0])
# indi = populationTu-pleSorted[0][0]
# yTotal = [0]*(401)
## lenY = 0
# for i in range(1,10):
# if i < 10:
# filename = front+'000'+str(i)+end
# elif i< 100:
# filename = front+'00'+str(i)+end
# else:
# filename = front+'0'+str(i)+end
# path=feffdat.feffpath(filename, s02=str(indi[i-1][0]), e0=str(indi[i-1][1]), sigma2=str(indi[i-1][2]), deltar=str(indi[i-1][3]), _larch=mylarch)
# feffdat._path2chi(path, _larch=mylarch)
# y = path.chi
## lenY = len(y)
# for k in range(len(y)):
# yTotal[k] += y[k]
# global g
## for m in range(lenY):
## yTotal[m] = yTotal[m]*g.k**2
# global global_yTotal
# plt.plot(g.k, g.chi*g.k**2)
# plt.plot(g.k[0:401], yTotal*g.k[0:401]**2)
## plt.ylim(top=6, bottom=-6)
# plt.show()
# file.write("Gen Num: %d" % genNum)
# file.write("Fitness:"+str(populationTupleSorted[0][1]))
# file.write("Combination:"+str(populationTupleSorted[0][0]))
nextBreeders = selectFromPopulation(populationSorted, best_sample, lucky_few)
nextPopulation = createChildren(nextBreeders, number_of_child)
lenDiff = len(firstGeneration)-len(nextPopulation)
for i in range(lenDiff):
j = random.randint(0,len(firstGeneration)-1)
nextPopulation.append(firstGeneration[j])
print(len(nextPopulation))
nextGeneration = mutatePopulation(nextPopulation, chance_of_mutation, chance_of_mutation_e0)
return nextGeneration
def multipleGeneration(number5_of_generation, exp, size_population, best_sample, lucky_few, number_of_child, chance_of_mutation, chance_of_mutation_e0):
historic = []
historic.append(generateFirstGen(size_population))
for i in range (number_of_generation):
historic.append(nextGeneration(historic[i], exp, best_sample, lucky_few, number_of_child, chance_of_mutation, chance_of_mutation_e0))
return historic
#printing tool - NOT DONE!!!!!!!!!!!!!
def printSimpleResult(historic, exp, number_of_generation): #bestSolution in historic. Caution not the last
result = getListBestIndividualFromHistorique(historic, exp)[number_of_generation-1]
print ("solution: \"" + result[0] + "\" de fitness: " + str(result[1]))
#analysis tools
def getBestIndividualFromPopulation (population, exp):
return computePerfPop(population, exp)[0]
def getListBestIndividualFromHistorique (historic, exp):
bestIndividuals = []
for population in historic:
bestIndividuals.append(getBestIndividualFromPopulation(population, exp))
return bestIndividuals
#for finding good e0
def findE0(bestFit):
listOfx = []
listOfy = []
bestFitList = [list(x) for x in bestFit]
for i in rangeB:
for j in bestFitList:
j[1] = i
fit = fitness(bestFitList)
listOfx.append(i)
listOfy.append(fit)
#main program
file = open("Result.txt","w")
g = read_ascii('Cu Data/cu_10k.xmu', _larch = mylarch)
autobk(g, rbkg=1.45, _larch = mylarch)
exp = g.chi
#kidNum = 0
genNum = 0
size_population = 1000
best_sample = 200
lucky_few = 200
number_of_child = 4
number_of_generation = 1000
chance_of_mutation = 20
chance_of_mutation_e0 = 0
historyBest = 0
bestDiff = 9999
diffCounter = 0
#e0
#b = random.choice(rangeB)
b = 1.86
if ((best_sample + lucky_few) / 2 * number_of_child >= size_population):
print ("population size not stable")
else:
historic = multipleGeneration(number_of_generation, exp, size_population, best_sample, lucky_few, number_of_child, chance_of_mutation, chance_of_mutation_e0)
printSimpleResult(historic, exp, number_of_generation)