forked from huangrqhub/istock
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ga2.py
723 lines (658 loc) · 31 KB
/
ga2.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
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
# -*- coding: utf-8 -*-
from __future__ import division
import math
import numpy as np
import pandas as pd
import random as rand
import talib as ta
import cx_Oracle
import tushare as ts
import copy
# import time
from datetime import datetime
import scipy.stats.mstats as mstats
from pymongo import MongoClient
def get_r(data):
x = [8,13,21,34,55]
cl = np.array(data['CL'],dtype='f8')
for i in x:
data['m'+str(i)] = ta.EMA(cl,timeperiod=i)
for i,r in data.iterrows():
if math.isnan(r['m55']): continue
z = [[r['m8'],8],[r['m13'],13],[r['m21'],21],[r['m34'],34],[r['m55'],55]]
c,p = mstats.spearmanr(x,[k[1] for k in sorted(z,reverse=True)])
data.loc[i,'r'] = c
return data['r']
#全局变量：
PSIZE=30#种群数量
TERMINAL=20#迭代代数
PC=0.95 #交叉概率
PM=0.2 #变异概率
#惩罚函数：
def get_stat(data):
data['deal'] = data['deal'].shift()
# data['deal'].iloc[(data['deal'].count()-1)] = 0
data['deal'].values[-1] = 0
cnt=len(data)
dt = data.dropna().copy()
if len(dt)==0: return [0,0,0,0]
dt['e'] = dt.OP/dt.OP.shift()*0.997
trcnt = dt['e'].count()
if trcnt==0: return [0,0,0,0]
es = dt.loc[dt['deal']==0]['e'].cumprod()
e=es.iloc[len(es)-1]
if math.isnan(e):
# print es
print 'length',len(es),cnt,trcnt
return [e*(1-trcnt/cnt), e, trcnt, cnt]
#双均线交叉：
def MA_Cross(price_dt,short_ma=10,long_ma=20):
s=ta.SMA(np.array(price_dt['CL'],dtype='f8'),timeperiod=short_ma)
l=ta.SMA(np.array(price_dt['CL'],dtype='f8'),timeperiod=long_ma)
signal=pd.DataFrame({'s':s,'l':l})
signal.loc[(signal['s']>signal['l']),'bs']=1
signal.loc[(signal['s']<signal['l']),'bs']=0
return signal['bs']
chromo_fun={'dema':ta.DEMA,'ema':ta.EMA,'kama':ta.KAMA,'midpoint':ta.MIDPOINT,'sma':ta.SMA,'tema':ta.TEMA,'trima':ta.TRIMA,'wma':ta.WMA,'macd':ta.MACD,'t3':ta.T3,'kdj':ta.STOCH,'mom':ta.MOM,'roc':ta.ROC,'rsi':ta.RSI,'trix':ta.TRIX,'ma_cross':MA_Cross,'ema_ali':get_r}
#指标计算：
def indic_cal(price_data,indic_dict):
chromo_fun={'dema':ta.DEMA,'ema':ta.EMA,'kama':ta.KAMA,'midpoint':ta.MIDPOINT,'sma':ta.SMA,'tema':ta.TEMA,'trima':ta.TRIMA,'wma':ta.WMA,'macd':ta.MACD,'t3':ta.T3,'kdj':ta.STOCH,'mom':ta.MOM,'roc':ta.ROC,'rsi':ta.RSI,'trix':ta.TRIX,'ma_cross':MA_Cross,'ema_ali':get_r}
indictor0=['dema','ema','kama','sma','tema','trima']#midpoint,wma被删除
for k in indic_dict:
if k in indictor0:
price_data[k]=chromo_fun[k](real=np.array(price_data['CL'],dtype='f8'),timeperiod=indic_dict[k]['parameter'])
price_data.loc[price_data[price_data['CL']>price_data[k]].index,k+'_SIGNAL']=1
price_data.loc[price_data[price_data['CL']<price_data[k]].index,k+'_SIGNAL']=0
if k=='ma_cross':
price_data['ma_cross_SIGNAL']=chromo_fun['ma_cross'](price_dt=price_data,short_ma=indic_dict['ma_cross']['parameter']['ma_cross_sma'],long_ma=indic_dict['ma_cross']['parameter']['ma_cross_lma'])
if k=='rsi':
price_data['rsi']=chromo_fun['rsi'](real=np.array(price_data['CL'],dtype='f8'),timeperiod=indic_dict['rsi']['parameter'])
price_data.loc[price_data[price_data['CL']>price_data['rsi']].index,'rsi_SIGNAL']=1
price_data.loc[price_data[price_data['CL']<price_data['rsi']].index,'rsi_SIGNAL']=0
if k=='ema_ali':
price_data['ema_ali']=chromo_fun['ema_ali'](price_data)
price_data.loc[price_data[0<price_data['ema_ali']].index,'ema_ali_SIGNAL']=1
price_data.loc[price_data[0>price_data['ema_ali']].index,'ema_ali_SIGNAL']=0
if k=='roc':
price_data['roc']=chromo_fun['roc'](real=np.array(price_data['CL'],dtype='f8'),timeperiod=indic_dict['roc']['parameter'])
price_data.loc[price_data[price_data['CL']>price_data['roc']].index,'roc_SIGNAL']=1
price_data.loc[price_data[price_data['CL']<price_data['roc']].index,'roc_SIGNAL']=0
if k=='mom':
price_data['mom']=chromo_fun['mom'](real=np.array(price_data['CL'],dtype='f8'),timeperiod=indic_dict['mom']['parameter'])
price_data.loc[price_data[price_data['CL']>price_data['mom']].index,'mom_SIGNAL']=1
price_data.loc[price_data[price_data['CL']<price_data['mom']].index,'mom_SIGNAL']=0
if k=='macd':
price_data['macd'],price_data['macdsignal'],price_data['macdhist']=chromo_fun['macd'](np.array(price_data['CL'],dtype='f8'))
price_data.loc[price_data[0<price_data['macd']].index,'DIFF']=1
price_data.loc[price_data[0<price_data['macdsignal']].index,'DEA']=1
price_data.loc[price_data[price_data['macd']>price_data['macdsignal']].index,'D']=1
price_data.loc[price_data[0>price_data['macd']].index,'DIFF']=0
price_data.loc[price_data[0>price_data['macdsignal']].index,'DEA']=0
price_data.loc[price_data[price_data['macd']<price_data['macdsignal']].index,'D']=0
price_data['macd_SIGNAL']=price_data['DIFF']*price_data['DEA']*price_data['D']
if k=='t3':
price_data['t3']=chromo_fun['t3'](real=np.array(price_data['CL'],dtype='f8'))
price_data.loc[price_data[price_data['CL']>price_data['t3']].index,'t3_SIGNAL']=1
price_data.loc[price_data[price_data['CL']<price_data['t3']].index,'t3_SIGNAL']=0
if k=='kdj':
price_data['kdj_slowk'],price_data['kdj_slowd']=chromo_fun['kdj'](np.array(price_data['HI'],dtype='f8'),np.array(price_data['LO'],dtype='f8'),np.array(price_data['CL'],dtype='f8'))
price_data.loc[price_data[price_data['kdj_slowk']>price_data['kdj_slowd']].index,'kdj'+'_SIGNAL']=1
price_data.loc[price_data[price_data['kdj_slowk']<price_data['kdj_slowd']].index,'kdj'+'_SIGNAL']=0
l=len(price_data)
signal=0
for k in indic_dict:
signal+=price_data[k+'_SIGNAL'].iloc[l-1]*indic_dict[k]['weight']
if signal>0.5:
return 1
else:
return 0
#股票代码获取函数：
def sc_get():
db = cx_Oracle.connect('mc4','mc4998','192.168.0.18:1521/racdb')
cur = db.cursor()
return cur.execute('SELECT stkcode FROM vw_stk_code').fetchall()
#数据获取函数：
def price_get(stock_code='300109'):
db = cx_Oracle.connect('mc4','mc4998','192.168.0.18:1521/racdb')
# sql = '''
# select trdate,v_open*(adj_price/v_close) op,v_high*(adj_price/v_close) hi,v_low*(adj_price/v_close) lo,adj_price cl,v_change ch
# from stk_mkt where stkcode='%s' order by trdate'''
sql = '''
select trdate,v_open*(adj_price/v_close) op,v_high*(adj_price/v_close) hi,v_low*(adj_price/v_close) lo,adj_price cl,v_change ch
from stk_mkt_ex_t where stkcode='%s' order by trdate
'''
data = pd.read_sql(sql % (stock_code,),db)
return data
#归一化函数：
def popu_norm(popu_mem):
for i in range(len(popu_mem)):
ep=popu_mem[i]
sumrate=sum(ep.chromosome.values())
ep.chromosome={k:ep.chromosome[k]/sumrate for k in chromo_name}
return popu_mem
#指标集：
class indic_w(object):
def __init__(self,thred=0.02):
self.chromo_n0=['dema','ema','kama','sma','tema','trima']#midpoint,wma被删除
self.chromo_n1=['mom','roc','rsi','ema_ali']
self.chromo_n2=['macd','t3','kdj','ma_cross']
self.chromo_n=self.chromo_n0+self.chromo_n1+self.chromo_n2
len_chname=len(self.chromo_n)
self.chromo0={i:1/len_chname for i in self.chromo_n}
self.thred=thred
self.dump=[]
self.fitness=0
self.opt_list=[]
self.idx=0
self.opt_para={}
self.opt_prof=[]
self.cn0=['dema','ema','kama','sma','tema','trima']
self.cn1=['mom','roc','rsi','ema_ali']
self.cn2=['macd','t3','kdj','ma_cross']
self.wei_dict={}
self.data_signal=[]
self.opt_gene={}
#淘汰函数
def eliminate(self,wei_dict,fitness):
if fitness<self.fitness:
if self.idx>=len(self.opt_list):
print('it is the best')
self.opt_gene={k:{'weight':self.wei_dict[k],'parameter':self.opt_para[k]} for k in self.wei_dict}
print('optimization',self.opt_gene)
return 0
if self.dump[-1] in self.cn0:
self.chromo_n0.append(self.dump[-1])
elif self.dump[-1] in self.cn1:
self.chromo_n1.append(self.dump[-1])
else:
self.chromo_n2.append(self.dump[-1])
self.opt_list.insert(self.idx,self.dump[-1])
self.dump.remove(self.dump[-1])
self.dump.append(self.opt_list[self.idx+1])
self.opt_list.remove(self.opt_list[self.idx+1])
self.idx+=1
if self.dump[-1] in self.chromo_n0:
self.chromo_n0.remove(self.dump[-1])
elif self.dump[-1] in self.chromo_n1:
self.chromo_n1.remove(self.dump[-1])
else:
self.chromo_n2.remove(self.dump[-1])
self.chromo_n=self.chromo_n0+self.chromo_n1+self.chromo_n2
len_chname=len(self.chromo_n)
self.chromo0={i:1/len_chname for i in self.chromo_n}
else:
self.fitness=fitness
self.wei_dict=wei_dict
wei_list=[[wei_dict[k],k] for k in wei_dict]
e_indic=sorted(wei_list)
self.opt_list=[j[1] for j in e_indic]
if e_indic[0][0]<self.thred:
if e_indic[0][1] in self.chromo_n0:
self.chromo_n0.remove(e_indic[0][1])
elif e_indic[0][1] in self.chromo_n1:
self.chromo_n1.remove(e_indic[0][1])
else:
if e_indic[0][1] not in self.chromo_n2:
print('indictor removed',e_indic[0][1])
print('chromo_n2',self.chromo_n2)
self.chromo_n2.remove(e_indic[0][1])
self.chromo_n=self.chromo_n0+self.chromo_n1+self.chromo_n2
len_chname=len(self.chromo_n)
self.chromo0={i:1/len_chname for i in self.chromo_n}
self.dump.append(e_indic[0][1])
self.idx=0
else:
print('all weight are bigger than threshold')
self.opt_gene={k:{'weight':self.wei_dict[k],'parameter':self.opt_para[k]} for k in self.wei_dict}
return 0
return 1
#优化参数
def para_opt(self,stock_code,period=[range(10,16)+range(45,61)]):
data=price_get(stock_code)
self.data_signal=copy.deepcopy(data)
bot=-10
for k in self.cn0:
opt_k=0
opt_p=bot
for p in period:
data[k]=chromo_fun[k](real=np.array(data['CL'],dtype='f8'),timeperiod=p)
data.loc[data[data['CL']>data[k]].index,k+'_SIGNAL']=1
data.loc[data[data['CL']<data[k]].index,k+'_SIGNAL']=0
if 'deal' in data.columns : del data['deal']
data.loc[data[data[k+'_SIGNAL']>data[k+'_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data[k+'_SIGNAL']<data[k+'_SIGNAL'].shift()].index,'deal']=0
ep=get_stat(data)[0]
if ep>opt_p:
opt_k,opt_p=p,ep
print(k,opt_k,opt_p)
if opt_p<bot:
self.chromo_n0.remove(k)
print(k,'is out')
continue
self.opt_para[k]=opt_k
self.opt_prof.append([opt_p,k])
self.data_signal[k]=chromo_fun[k](real=np.array(self.data_signal['CL'],dtype='f8'),timeperiod=opt_k)
self.data_signal.loc[self.data_signal[self.data_signal['CL']>self.data_signal[k]].index,k+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['CL']<self.data_signal[k]].index,k+'_SIGNAL']=0
#均线交叉
opt_ks=0
opt_kl=0
opt_p=bot
for ps in range(7,120,7):
pl=ps+7
while(pl<140):
data['ma_cross_SIGNAL']=MA_Cross(price_dt=data,short_ma=ps,long_ma=pl)
if 'deal' in data.columns : del data['deal']
data.loc[data[data[k+'_SIGNAL']>data[k+'_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data[k+'_SIGNAL']<data[k+'_SIGNAL'].shift()].index,'deal']=0
ep=get_stat(data)[0]
if ep>opt_p:
opt_ks,opt_kl,opt_p=ps,pl,ep
pl+=7
# self.opt_para['ma_cross']={'ma_cross_sma':opt_ks,'ma_cross_lma':opt_kl}
# self.opt_prof.append([opt_p,'ma_cross'])
print('ma_cross',opt_ks,opt_kl,opt_p)
if opt_p<bot:
self.chromo_n2.remove('ma_cross')
print('ma_cross is out')
else:
self.opt_para['ma_cross']={'ma_cross_sma':opt_ks,'ma_cross_lma':opt_kl}
self.opt_prof.append([opt_p,'ma_cross'])
self.data_signal['ma_cross_SIGNAL']=chromo_fun['ma_cross'](price_dt=data,short_ma=opt_ks,long_ma=opt_kl)
#rsi指标
opt_k=0
opt_p=bot
for p in period:
data['rsi']=chromo_fun['rsi'](np.array(data['CL'],dtype='f8'),timeperiod=p)
data.loc[data[50<data['rsi']].index,'rsi'+'_SIGNAL']=1
data.loc[data[50>data['rsi']].index,'rsi'+'_SIGNAL']=0
if 'deal' in data.columns : del data['deal']
data.loc[data[data['rsi_SIGNAL']>data['rsi_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['rsi_SIGNAL']<data['rsi_SIGNAL'].shift()].index,'deal']=0
ep=get_stat(data)[0]
if ep>opt_p:
opt_k,opt_p=p,ep
print('rsi',opt_k,opt_p)
if opt_p<bot:
self.chromo_n1.remove('rsi')
print('rsi is out')
else:
self.opt_para['rsi']=opt_k
self.opt_prof.append([opt_p,'rsi'])
self.data_signal['rsi']=chromo_fun['rsi'](real=np.array(self.data_signal['CL'],dtype='f8'),timeperiod=opt_k)
self.data_signal.loc[self.data_signal[self.data_signal['CL']>self.data_signal['rsi']].index,'rsi'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['CL']<self.data_signal['rsi']].index,'rsi'+'_SIGNAL']=0
#多头排列指标
opt_p=bot
self.data_signal['ema_ali']=chromo_fun['ema_ali'](copy.deepcopy(data))
self.data_signal.loc[self.data_signal[0<self.data_signal['ema_ali']].index,'ema_ali'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[0>self.data_signal['ema_ali']].index,'ema_ali'+'_SIGNAL']=0
data['ema_ali_SIGNAL']=self.data_signal['ema_ali_SIGNAL']
if 'deal' in data.columns : del data['deal']
data.loc[data[data['ema_ali_SIGNAL']>data['ema_ali_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['ema_ali_SIGNAL']<data['ema_ali_SIGNAL'].shift()].index,'deal']=0
opt_p=get_stat(data)[0]
print('ema_ali',opt_p)
if opt_p>bot:
self.opt_para['ema_ali']={'parameter':'default'}
self.opt_prof.append([opt_p,'ema_ali'])
else:
self.chromo_n1.remove('ema_ali')
print('ema_ali is out')
#roc指标
opt_k=0
opt_p=bot
for p in period:
data['roc']=chromo_fun['roc'](np.array(data['CL'],dtype='f8'),timeperiod=p)
data.loc[data[0<data['roc']].index,'roc'+'_SIGNAL']=1
data.loc[data[0>data['roc']].index,'roc'+'_SIGNAL']=0
if 'deal' in data.columns : del data['deal']
data.loc[data[data['roc_SIGNAL']>data['roc_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['roc_SIGNAL']<data['roc_SIGNAL'].shift()].index,'deal']=0
ep=get_stat(data)[0]
if ep>opt_p:
opt_k,opt_p=p,ep
print('roc',opt_k,opt_p)
if opt_p<bot:
self.chromo_n1.remove('roc')
print('roc is out')
else:
self.opt_para['roc']=opt_k
self.opt_prof.append([opt_p,'roc'])
self.data_signal['roc']=chromo_fun['roc'](real=np.array(self.data_signal['CL'],dtype='f8'),timeperiod=opt_k)
self.data_signal.loc[self.data_signal[self.data_signal['CL']>self.data_signal['roc']].index,'roc'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['CL']<self.data_signal['roc']].index,'roc'+'_SIGNAL']=0
#mom指标
opt_k=0
opt_p=bot
for p in period:
data['mom']=chromo_fun['mom'](np.array(data['CL'],dtype='f8'),timeperiod=p)
data.loc[data[0<data['mom']].index,'mom'+'_SIGNAL']=1
data.loc[data[0>data['mom']].index,'mom'+'_SIGNAL']=0
if 'deal' in data.columns : del data['deal']
data.loc[data[data['mom_SIGNAL']>data['mom_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['mom_SIGNAL']<data['mom_SIGNAL'].shift()].index,'deal']=0
ep=get_stat(data)[0]
if ep>opt_p:
opt_k,opt_p=p,ep
print('mom',opt_k,opt_p)
if opt_p<bot:
self.chromo_n1.remove('mom')
print('mom is out')
else:
self.opt_para['mom']=opt_k
self.opt_prof.append([opt_p,'mom'])
self.data_signal['mom']=chromo_fun['mom'](real=np.array(self.data_signal['CL'],dtype='f8'),timeperiod=opt_k)
self.data_signal.loc[self.data_signal[self.data_signal['CL']>self.data_signal['mom']].index,'mom'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['CL']<self.data_signal['mom']].index,'mom'+'_SIGNAL']=0
#macd指标
opt_p=bot
self.data_signal['macd'],self.data_signal['macdsignal'],self.data_signal['macdhist']=chromo_fun['macd'](np.array(self.data_signal['CL'],dtype='f8'))
self.data_signal.loc[self.data_signal[0<self.data_signal['macd']].index,'DIFF']=1
self.data_signal.loc[self.data_signal[0<self.data_signal['macdsignal']].index,'DEA']=1
self.data_signal.loc[self.data_signal[self.data_signal['macd']>self.data_signal['macdsignal']].index,'D']=1
self.data_signal.loc[self.data_signal[0>self.data_signal['macd']].index,'DIFF']=0
self.data_signal.loc[self.data_signal[0>self.data_signal['macdsignal']].index,'DEA']=0
self.data_signal.loc[self.data_signal[self.data_signal['macd']<self.data_signal['macdsignal']].index,'D']=0
self.data_signal['macd_SIGNAL']=self.data_signal['DIFF']*self.data_signal['DEA']*self.data_signal['D']
data['macd_SIGNAL']=self.data_signal['macd_SIGNAL']
if 'deal' in data.columns : del data['deal']
data.loc[data[data['macd_SIGNAL']>data['macd_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['macd_SIGNAL']<data['macd_SIGNAL'].shift()].index,'deal']=0
opt_p=get_stat(data)[0]
print('macd',{'fastperiod':12, 'slowperiod':26, 'signalperiod':9},opt_p)
if opt_p>bot:
self.opt_para['macd']={'fastperiod':12, 'slowperiod':26, 'signalperiod':9}
self.opt_prof.append([opt_p,'macd'])
else:
self.chromo_n2.remove('macd')
print('macd is out')
#T3指标
opt_p=bot
self.data_signal['t3']=chromo_fun['t3'](np.array(self.data_signal['CL'],dtype='f8'))
self.data_signal.loc[self.data_signal[self.data_signal['CL']>self.data_signal['t3']].index,'t3'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['CL']<self.data_signal['t3']].index,'t3'+'_SIGNAL']=0
data['t3_SIGNAL']=self.data_signal['t3_SIGNAL']
if 'deal' in data.columns : del data['deal']
data.loc[data[data['t3_SIGNAL']>data['t3_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['t3_SIGNAL']<data['t3_SIGNAL'].shift()].index,'deal']=0
opt_p=get_stat(data)[0]
print('t3',{'timeperiod':5, 'vfactor':0.7},opt_p)
if opt_p>bot:
self.opt_para['t3']={'timeperiod':5, 'vfactor':0.7}
self.opt_prof.append([opt_p,'t3'])
else:
self.chromo_n2.remove('t3')
print('t3 is out')
#kdj指标
opt_p=bot
self.data_signal['kdj_slowk'],self.data_signal['kdj_slowd']=ta.STOCH(np.array(self.data_signal['HI'],dtype='f8'),np.array(self.data_signal['LO'],dtype='f8'),np.array(self.data_signal['CL'],dtype='f8'))
self.data_signal.loc[self.data_signal[self.data_signal['kdj_slowk']>self.data_signal['kdj_slowd']].index,'kdj'+'_SIGNAL']=1
self.data_signal.loc[self.data_signal[self.data_signal['kdj_slowk']<self.data_signal['kdj_slowd']].index,'kdj'+'_SIGNAL']=0
data['kdj_SIGNAL']=self.data_signal['kdj_SIGNAL']
if 'deal' in data.columns : del data['deal']
data.loc[data[data['kdj_SIGNAL']>data['kdj_SIGNAL'].shift()].index,'deal']=1
data.loc[data[data['kdj_SIGNAL']<data['kdj_SIGNAL'].shift()].index,'deal']=0
opt_p=get_stat(data)[0]
print('kdj',{'fastk_period':5, 'slowk_period':3, 'slowk_matype':0, 'slowd_period':3, 'slowd_matype':0},opt_p)
if opt_p>bot:
self.opt_para['kdj']={'fastk_period':5, 'slowk_period':3, 'slowk_matype':0, 'slowd_period':3, 'slowd_matype':0}
self.opt_prof.append([opt_p,'kdj'])
else:
self.chromo_n2.remove('kdj')
print('kdj is out')
self.chromo_n=self.chromo_n0+self.chromo_n1+self.chromo_n2
# def indic_bs(self):
#评价体系类：
class eval_sys(object):
def __init__(self,prof=0,win_pro=0,match_rate=0,sharp=0,retrace_max=0,deal_times=0):
self.prof=prof
self.win_pro=win_pro
self.match_rate=match_rate
self.sharp=sharp
self.retrace_max=retrace_max
self.deal_times=deal_times
self.fitscore=0
def score_cal(self):pass
#种群类：
class popu(object):
def __init__(self,chromosome,amount=0.0,sharp_rate=0.0,win_rate=0.0,fitness=0.0):
self.chromosome=chromosome#指标权重参数字典
self.amount=amount#资金总量
self.sharp_rate=sharp_rate
self.win_rate=win_rate
self.fitness=fitness#适应度
#初始化种群：
def popu_init(chromo_name_i,chromo_i,style=0,chn=[]):
if style==0:
popu_mem=[popu(chromosome=chromo_i) for i in range(PSIZE-len(chromo_i))]
elif style==1:
popu_mem=[popu(chromosome=chromo_i[j]) for j in range(len(chromo_i))]
else:
popu_mem=[popu(chromosome={j:rand.uniform(0,1) for j in chromo_name_i}) for i in range(PSIZE)]
for i in range(PSIZE):
s=sum(popu_mem[i].chromosome.values())
for j in range(len(chromo_name_i)):
popu_mem[i].chromosome[chromo_name_i[j]]/=s
return popu_mem
#解码函数：
def decode(popu_mem,chnd0,chnd1,chnd2,chnd,data_sig,para_d):
if popu_mem[0].fitness==0:
sta=0
else:
sta=PSIZE
for i in range(sta,len(popu_mem)):
popu_mem[i].fitness=fitness_cal(indic_wei0=popu_mem[i].chromosome,chn0=chnd0,chn1=chnd1,chn2=chnd2,chn=chnd,para=para_d,data=copy.deepcopy(data_sig))
return popu_mem
#交叉函数：
def crossover(popu_mem,chromo_name_c):
len_p=len(popu_mem)
len_chn=len(chromo_name_c)
for i in range(0,len_p,2):
if rand.uniform(0,1)<PC:
temp_chromo1={}
temp_chromo2={}
cr_pos=rand.randint(0,len_chn-2)
x=i
y=i+1
if rand.randint(0,1)==1:
x=i+1
y=i
for j in range(len_chn):
if cr_pos<j:
temp_chromo1[chromo_name_c[j]]=popu_mem[x].chromosome[chromo_name_c[j]]
temp_chromo2[chromo_name_c[j]]=popu_mem[y].chromosome[chromo_name_c[j]]
else:
temp_chromo1[chromo_name_c[j]]=popu_mem[y].chromosome[chromo_name_c[j]]
temp_chromo2[chromo_name_c[j]]=popu_mem[x].chromosome[chromo_name_c[j]]
popu_mem.append(popu(chromosome=temp_chromo1))
popu_mem.append(popu(chromosome=temp_chromo2))
return popu_mem
#变异函数：
def mutation(popu_mem,chromo_name_m):
for i in range(PSIZE):
len_chname=len(chromo_name_m)
if rand.uniform(0,1)<PC:
temp_chromo=copy.copy(popu_mem[i].chromosome)
mt_pos=rand.randint(0,len_chname-1)
temp_chromo[chromo_name_m[mt_pos]]=rand.uniform(0,1)
popu_mem.append(popu(chromosome=temp_chromo))
return popu_mem
#适应度函数：
def fitness_cal(indic_wei0,chn0,chn1,chn2,chn,para,data=price_get(),dealtimes_need=0,thred_b=0.5,thred_s=0.5):
#归一化
sf=sum(indic_wei0.values())
if sf==0:
return 0
temp_chf={k:indic_wei0[k]/sf for k in indic_wei0}
indic_wei=temp_chf
for each_name in chn:
data[each_name+'_SIGNAL']=data[each_name+'_SIGNAL']*indic_wei[each_name]
data['wei_indic']=0
for k in chn:
data['wei_indic']=data['wei_indic']+data[k+'_SIGNAL']
data.loc[data[data['wei_indic']>thred_b].index,'bs']=1
data.loc[data[data['wei_indic']<thred_s].index,'bs']=0
data.loc[data[data['bs']>data['bs'].shift()].index,'deal']=1
data.loc[data[data['bs']<data['bs'].shift()].index,'deal']=0
data['deal']=data['deal'].shift()
cnt=len(data)
dt=data.dropna().copy()
if len(dt)==0:return 0
trcnt=len(dt)
dt['net_prof']=dt.OP/dt.OP.shift()*0.997
es=dt.loc[dt['bs']==0,'net_prof'].cumprod()
if len(es)==0:return 0
e=es.iloc[len(es)-1]
if dealtimes_need>0:
print('deal_days',len(data['TRDATE']))
print('deal_times',len(data.loc[data[(data['bs']<>data['bs'].shift(1))].index]))
print('prof in fact',e)
print('fitness',e*(1-trcnt/cnt))
return e
# if math.isnan(e): print('fit_cal_nan',e)
return e*(1-trcnt/cnt)
#累积适应度函数：
def sumfit(fitness_list):
sumf=fitness_list[:1]
# print('sumf_all',sumf)
for i in fitness_list:
if math.isnan(i):
# print('fitness',i,fitness_list.index(i))
break
for i in range(len(fitness_list))[1:]:
sumf.append(sumf[-1]+fitness_list[i])
if math.isnan(sumf[i]):
print('sumf',i,sumf[i],sumf[0])
break
return list(np.array(sumf)/sumf[-1])
#种群适应度排序函数：
def popu_sort(popu_mem):
lp=len(popu_mem)
for i in range(lp):
if math.isnan(popu_mem[i].fitness):
popu_mem[i].fitness=0
fit_idx=[[popu_mem[i].fitness,i] for i in range(lp)]
fit_sort=sorted(fit_idx)
popu_new=[popu_mem[fit_sort[i][1]] for i in range(lp)]
for i in range(lp):
if math.isnan(popu_new[i].fitness):
print('popu_sort_nan',i,popu_new[i].fitness)
return popu_new
#选择函数：
def select(popu_mem):
if len(popu_mem)<PSIZE:
print('popu size',len(popu_mem))
fl=sumfit([i.fitness for i in popu_mem])#适应度累积分布
if len(fl)<len(popu_mem):
print('fl len',len(fl),'popu len',len(popu_mem))
fl_idx=[[fl[i],i] for i in range(len(popu_mem))]
temp_popu=[]
ps=sorted([[rand.uniform(0,1)] for i in range(PSIZE)])#随机选择数列表
idx=0
j=0
while(idx<PSIZE):
if j>=len(popu_mem):
print('j',j,'fl[j-1]',fl[j-1],'i',idx,'ps(i)',ps[idx])
if fl[j]>=ps[idx]:
temp_popu.append(popu_mem[j])
idx+=1
else:
j+=1
#精英机制：确保留下原种群中适应度最大的
if temp_popu[-1].fitness<popu_mem[-1].fitness:
temp_popu[-1]=popu_mem[-1]
return temp_popu
#主函数：
def GA_main(init0=0,sc_ga='300109'):
spt=1
# 初始化种群：
flag=1
indic_w0=indic_w(thred=0.10)
#优化参数：
indic_w0.para_opt(stock_code=sc_ga,period=range(20,60))
# t=time.time()
while(flag>0):
chromo_name0=indic_w0.chromo_n0
chromo_name1=indic_w0.chromo_n1
chromo_name2=indic_w0.chromo_n2
chromo_name=indic_w0.chromo_n
len_chname=len(chromo_name)
#等权重基因初始化：
chromo0={j:1/len_chname for j in chromo_name}#每个指标（染色体）的初始权重
#按最优化参数所得收益进行归一化，即指标的初始权重等于该指标最优化收益率占所有指标最优化收益率总和的百分率
for k in indic_w0.opt_prof:
if k[0]<0:
k[0]=0.005
opt_prof_dict={k[1]:k[0] for k in indic_w0.opt_prof}
prof_opt=[opt_prof_dict[k] for k in indic_w0.chromo_n]
sumopt=sum(prof_opt)
chromo_init_fit={k:opt_prof_dict[k]/sumopt for k in indic_w0.chromo_n}
#按照仅使用单个指标进行初始化
single_chromo_popu=[]
single_chromo={k:0 for k in indic_w0.chromo_n}
single_i=0
for k in single_chromo:
temp_single=copy.deepcopy(single_chromo)
temp_single[k]=1
single_chromo_popu.append(temp_single)
single_i+=1
popu_mem01=popu_init(chromo_name_i=chromo_name,style=init0,chromo_i=chromo_init_fit)
popu_mem02=popu_init(chromo_name_i=chromo_name,style=1,chromo_i=single_chromo_popu)
popu_mem0=popu_mem01+popu_mem02
i=0
while(i<=TERMINAL):
s=sum(popu_mem0[-1].chromosome.values())
temp_ch={k:popu_mem0[-1].chromosome[k]/s for k in popu_mem0[-1].chromosome}
#交叉
popu_mem1=crossover(popu_mem0,chromo_name_c=chromo_name)
#变异
popu_mem2=mutation(popu_mem1,chromo_name_m=chromo_name)
#解码
popu_mem3=decode(popu_mem=popu_mem2,chnd0=chromo_name0,chnd1=chromo_name1,chnd2=chromo_name2,chnd=chromo_name,data_sig=indic_w0.data_signal,para_d=indic_w0.opt_para)
#排序
popu_mem4=popu_sort(popu_mem3)
#选择
popu_mem0=select(popu_mem4)
i+=1
sf=sum(popu_mem0[-1].chromosome.values())
temp_chf={k:popu_mem0[-1].chromosome[k]/sf for k in popu_mem0[-1].chromosome}
#淘汰
flag=indic_w0.eliminate(wei_dict=temp_chf,fitness=popu_mem0[-1].fitness)
if flag==0:
fitness_cal(indic_wei0=indic_w0.wei_dict,chn0=chromo_name0,chn1=chromo_name1,chn2=chromo_name2,chn=indic_w0.wei_dict.keys(),para=indic_w0.opt_para,data=copy.deepcopy(indic_w0.data_signal),dealtimes_need=1)
return {'profit':indic_w0.fitness,'opt_indic':copy.copy(indic_w0.opt_gene)}
#client = MongoClient('mongodb://mcdb:mcdb@192.168.1.142:27017/mcdb')
#mdb=client.mcdb
#c=mdb.mx
def get_result(sci):
# a=sc_get()
# for sci in range(len(a['STOCKCODE'])):
# a['STOCKCODE'][sci]='300344'
print('stock',sci,'starting')
data=price_get(stock_code=sci)
if len(data)<100:
print('可交易天数少于60，不予处理')
return
res=GA_main(init0=0,sc_ga=sci)
res.update({'stock_code':sci,'lasttime':datetime.now()})
print('result',res)
return res
# c.save(res)
def md_read(code0):
client = MongoClient('mongodb://mcdb:mc969@192.168.0.32:27017/mcdb')
mdb = client.mcdb
c = mdb.ga
print(type(c))
print(c.collection_names)
for code in code0:
dict_zb2 = c.find_one({'stock_code':code})
print(dict_zb2)
if __name__ == "__main__":
print 'starting at:',datetime.now()
import sys
h=get_result(sys.argv[1])
print(h)
print(type(h))
print 'end at:',datetime.now()
indic_cal(price_data=price_get(stock_code='000004'),indic_dict=h['opt_indic'])