/
geneModule.py
365 lines (324 loc) · 12.9 KB
/
geneModule.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
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
#Generates and breeds cubes
import CubeGrabbingWorld, CubeFightingWorld
import random
import operator
import pickle
import pylab
class Cube:
def __init__(self, strat):
self.dna = strat
self.fitness = 0
self.replay = []
def getFitness(self):
return self.fitness
def simulate(self, time = 1000):
result = CubeGrabbingWorld.simulateIndividual(self.dna, False, '', time = time)
self.fitness = result[0]
return self.fitness
def simulateAndRecord(self, f= '', time = 1000):
result = CubeGrabbingWorld.simulateIndividual(self.dna, display = False, recordThis = True, filename = f, time = time)
self.fitness = result[0]
self.replay = result[1]
return self.fitness
def play(self):
CubeGrabbingWorld.showReplay(self.replay)
def saveReplay(self, filename):
filename = filename + '.replay'
replayFile = open(filename, 'wb')
pickle.dump(self.replay, replayFile)
replayFile.close()
def saveDna(self, filename):
filename = filename + '.dna'
dnaFile = open(filename, 'wb')
pickle.dump(self.dna, dnaFile)
dnaFile.close()
def loadDna(self, filename):
filename = filename + '.dna'
dnaFile = open(filename, 'rb')
self.dna = pickle.load(dnaFile)
dnaFile.close()
def mutateDna(self, rate):
mutate(rate, self.dna)
def saveReplay2(replay, filename):
filename = filename + '.replay2'
replayFile = open(filename, 'wb')
pickle.dump(replay, replayFile)
replayFile.close()
def mutate(rate, dna):
for i in range(CubeGrabbingWorld.g1Scale):
for j in range(CubeGrabbingWorld.g2Scale):
for k in range(CubeGrabbingWorld.g3Scale):
if rate > random.random():
dna[i][j][k] = random.randint(1, CubeGrabbingWorld.actions)
def mutate2(rate, dna):
for i in range(CubeFightingWorld.gScale[0]):
for j in range(CubeFightingWorld.gScale[1]):
for k in range(CubeFightingWorld.gScale[2]):
for l in range(CubeFightingWorld.gScale[3]):
for m in range(CubeFightingWorld.gScale[4]):
if rate > random.random():
dna[i][j][k][l][m] = random.randint(1, CubeFightingWorld.actions)
def createChild(parent1, parent2, crossoverPoints, mutationRate):
spots = []
dnaFromParent1 = True
newDna = []
count = 0
for i in range(crossoverPoints):
spots.append(random.randint(0, CubeGrabbingWorld.g1Scale * CubeGrabbingWorld.g2Scale*CubeGrabbingWorld.g3Scale))
for i in range(CubeGrabbingWorld.g1Scale):
newDna.append([])
for j in range(CubeGrabbingWorld.g2Scale):
newDna[i].append([])
for k in range(CubeGrabbingWorld.g3Scale):
if count in spots:
dnaFromParent1 = not dnaFromParent1
if dnaFromParent1:
newDna[i][j].append(parent1.dna[i][j][k])
else:
newDna[i][j].append(parent2.dna[i][j][k])
count = count + 1
mutate(mutationRate, newDna)
return Cube(newDna)
def createChild2(parent1, parent2, crossoverPoints, mutationRate):
spots = []
dnaFromParent1 = True
newDna = []
count = 0
for i in range(crossoverPoints):
spots.append(random.randint(0, reduce(lambda x,y:x*y,CubeFightingWorld.gScale)))
for i in range(CubeFightingWorld.gScale[0]):
newDna.append([])
for j in range(CubeFightingWorld.gScale[1]):
newDna[i].append([])
for k in range(CubeFightingWorld.gScale[2]):
newDna[i][j].append([])
for l in range(CubeFightingWorld.gScale[3]):
newDna[i][j][k].append([])
for m in range(CubeFightingWorld.gScale[4]):
#print newDna
if count in spots:
dnaFromParent1 = not dnaFromParent1
if dnaFromParent1:
newDna[i][j][k][l].append(parent1.dna[i][j][k][l][m])
else:
newDna[i][j][k][l].append(parent2.dna[i][j][k][l][m])
count = count + 1
mutate2(mutationRate, newDna)
return Cube(newDna)
def getRandomDNA():
strat = [[[random.randint(1, 3) for i in range(10)] for j in range(10)] for k in range(10)]
return strat
def getRandomDNA2():
strat = [[[[[random.randint(1, 3) for i in range(10)] for j in range(10)] for k in range(10)] for l in range(10)] for m in range(10)]
return strat
def playReplay(filename):
CubeGrabbingWorld.showReplay(filename)
def playReplay2(filename):
CubeFightingWorld.showReplay(filename)
def sortPopulation(population):
population = sorted(population, key=operator.attrgetter('fitness'))
population.reverse()
return population
def selectFromFitnessWheel(population, sum):
value = random.randint(1, sum)
sum = 0
for i in range(len(population)):
sum = sum + population[i].fitness
if value <= sum:
return population[i]
def getChildren(population, crossoverPoints, mutationRate):
newPop = []
sortPopulation(population)
sum = 0
for i in range(len(population)):
sum = sum + population[i].fitness
for i in range(len(population) - 1):
p1 = selectFromFitnessWheel(population, sum)
p2 = selectFromFitnessWheel(population, sum)
newPop.append(createChild(p1, p2, crossoverPoints, mutationRate))
population[0].fitness = 0
newPop.append(population[0])
return newPop
def getChildren2(population, crossoverPoints, mutationRate):
newPop = []
sortPopulation(population)
sum = 0
for i in range(len(population)):
sum = sum + population[i].fitness
for i in range(len(population) - 1):
p1 = selectFromFitnessWheel(population, sum)
p2 = selectFromFitnessWheel(population, sum)
newPop.append(createChild2(p1, p2, crossoverPoints, mutationRate))
population[0].fitness = 0
newPop.append(population[0])
return newPop
def getNewPopulation(size):
pop = []
for i in range(size):
pop.append(Cube(getRandomDNA()))
return pop
def getNewPopulation2(size):
pop = []
for i in range(size):
pop.append(Cube(getRandomDNA2()))
return pop
def simulatePopulation(pop, time = 1000):
for agent in pop:
agent.simulateAndRecord("", time)
def recordPopulation(pop, filePrefix, gen, doPrint = True, doRecord = True, gb = 0):
sum = 0
for i in range(len(pop)):
sum = sum + pop[i].fitness
avg = float(sum)/(len(pop))
if doPrint:
print("Generation " + str(gen) + ":")
print("Best has fitness of " + str(pop[0].fitness))
print("Average fitness is " + str(avg))
print("Best so far is " + str(gb))
if doRecord:
pop[0].saveReplay(filePrefix + "_gen_" + str(gen))
pop[0].saveDna(filePrefix + ' gen ' + str(gen))
return avg
def generationDetailPrintout(pop):
for i in range(len(pop)):
print("Agent " + str(i) + ": " + str(pop[i].fitness))
def savePlot(history, filePrefix):
for i in range(2):
pylab.plot(range(len(history)), extractNth(history, i))
pylab.xlabel('Generation')
pylab.ylabel('Fitness')
pylab.title(filePrefix + ' Fitness Plot')
pylab.savefig(filePrefix + '_plot.png')
# def savePlot(filePrefix):
# histFile = open(filePrefix + '.history', 'rb')
# history = pickle.load(histFile)
# for i in range(2):
# pylab.plot(range(len(history)), extractNth(history, i))
# pylab.xlabel('Generation')
# pylab.ylabel('Fitness')
# pylab.title(filePrefix + ' Fitness Plot')
# pylab.savefig(filePrefix + '_plot.png')
def savePlot2(history, filePrefix):
pylab.plot(range(len(history)), extractNth(history, 0), 'b')
pylab.plot(range(len(history)), extractNth(history, 1), 'b')
pylab.plot(range(len(history)), extractNth(history, 2), 'r')
pylab.plot(range(len(history)), extractNth(history, 3), 'r')
pylab.xlabel('Generation')
pylab.ylabel('Fitness')
pylab.title(filePrefix + ' Fitness Plot')
pylab.savefig(filePrefix + '_plot.png')
def savePlot2(filePrefix):
histFile = open(filePrefix + '.history2', 'rb')
history = pickle.load(histFile)
pylab.plot(range(len(history)), extractNth(history, 0), 'b')
pylab.plot(range(len(history)), extractNth(history, 1), 'b')
pylab.plot(range(len(history)), extractNth(history, 2), 'r')
pylab.plot(range(len(history)), extractNth(history, 3), 'r')
pylab.xlabel('Generation')
pylab.ylabel('Fitness')
pylab.title(filePrefix + ' Fitness Plot')
pylab.savefig(filePrefix + '_plot.png')
def runSimulation(generations, popSize, crossoverPoints, mutationRate, filePrefix, printEvery = 1, recordEvery = 1, time = 1000):
history = []
globalBest = 0
globalBestBot = 0
pop = getNewPopulation(popSize)
for i in range(generations):
simulatePopulation(pop, time)
pop = sortPopulation(pop)
best = pop[0].fitness
worst = pop[len(pop) - 1].fitness
if i % printEvery == 0:
p = True
#generationDetailPrintout(pop)
else:
p = False
if i % recordEvery == 0:
r = True
else:
r= False
if best > globalBest:
globalBest = best
globalBestBot = pop[0]
#r=True
#p=True
avg = recordPopulation(pop, filePrefix, i, p, r, globalBest)
history.append((best, avg))
pop = getChildren(pop, crossoverPoints, mutationRate)
if p:
savePlot(history, filePrefix)
globalBestBot.saveDna(filePrefix + 'Best')
globalBestBot.saveReplay(filePrefix + 'Best')
simulatePopulation(pop, time)
sortPopulation(pop)
recordPopulation(pop, filePrefix, i, True, True)
globalBestBot.saveDna(filePrefix + 'Best')
globalBestBot.saveReplay(filePrefix + 'Best')
savePlot(history, filePrefix)
return globalBestBot
def extractNth(seq, index):
return map(lambda x : x[index], seq)
def getAverage(bot, trials, t = 1000):
sum = 0
for i in range(trials):
bot.simulate(t)
sum = sum + bot.fitness
return float(sum) / trials
def ffSimple(a, b):
return (a, b)
def ffRed(a, b):
return(b, b)
def ffTeam(a, b):
return(a+b, a+b)
def compete(p1, p2, display = False, recordThis = False, filename = '', time = 1000, fitnessFunction = ffSimple):
result = CubeFightingWorld.simulateCompetition(p1.dna, p2.dna, display, recordThis, filename, time)
fits = fitnessFunction(result[0], result[1])
p1.fitness = p1.fitness + fits[0]
p2.fitness = p2.fitness + fits[1]
return (fits[0], fits[1], result[2])
def getHistoryPoint(pop1, pop2):
sum1 = 0
sum2 = 0
for bot in pop1:
sum1 = sum1 + bot.fitness
for bot in pop2:
sum2 = sum2 + bot.fitness
return (float(pop1[0].fitness)/len(pop2), float(sum1)/(len(pop2)*len(pop1)), float(pop2[0].fitness)/len(pop1), float(sum2)/(len(pop1)*len(pop2)))
def runCompetition(generations, popSize1, popSize2, crossoverPoints, mutationRate, filePrefix, printEvery = 10, recordEvery = 100, time = 1000, fitnessFunction = ffSimple):
history = []
globalBest = 0
globalBestReplay = 0
printThis = False
pop1 = getNewPopulation2(popSize1)
pop2 = getNewPopulation2(popSize2)
for i in range(generations):
printThis = i % printEvery == 0
for j in range(popSize1):
for k in range(popSize2):
temp = compete(pop1[j], pop2[k], False, True, '', time, fitnessFunction)
if temp[0] > globalBest or temp[1] > globalBest:
saveReplay2(temp[2], filePrefix + '_gen_' + str(i))
globalBest = max(temp[0], temp[1])
globalBestReplay = temp[2]
printThis = True
pop1 = sortPopulation(pop1)
pop2 = sortPopulation(pop2)
point = getHistoryPoint(pop1, pop2)
history.append(point)
if printThis:
print("Generation " + str(i))
print("Best blue: " + str(point[0]))
print("Average blue: " + str(point[1]))
print("Best red: " + str(point[2]))
print("Average red: " + str(point[3]))
generationDetailPrintout(pop1)
generationDetailPrintout(pop2)
pop1 = getChildren2(pop1, crossoverPoints, mutationRate)
pop2 = getChildren2(pop2, crossoverPoints, mutationRate)
saveReplay2(globalBestReplay, filePrefix + 'Best')
def main():
bot = runSimulation(200, 100, 3, .01, 'soloTest', 10, 100)
playReplay('soloTestBest')
#runCompetition(100, 10, 10, 3, .04, 'teamworkTest', 1, 1, time = 750, fitnessFunction = ffRed)
#playReplay2('teamworkTestBest')
main()