-
Notifications
You must be signed in to change notification settings - Fork 35
/
Copy pathv.0.4.2-framework.py
506 lines (391 loc) · 15.5 KB
/
v.0.4.2-framework.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
#v.0.4.x FULL IMPLEMENTATION (NN MODEL AND TRADING APP)
#v.0.4.1 portabilty (logging+config for exe)
#v.0.4.2 support for DukascopyBinary 1.10.26
import logging
import logging.config
logerrors = 0
if(logerrors):
logging.basicConfig(filename='tmp/error.log', level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger=logging.getLogger(__name__)
else:
print '\nconsole printing\n'
try:
import pandas as pd
import numpy as np
import random
import matplotlib.pylab as plt
import datetime
import locale
locale.setlocale(locale.LC_NUMERIC, "")
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation, Flatten
from keras.layers.recurrent import LSTM, GRU
from keras.layers import Convolution1D, MaxPooling1D, AtrousConvolution1D, RepeatVector
from keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, CSVLogger
from keras.layers.wrappers import Bidirectional
from keras import regularizers
from keras.layers.normalization import BatchNormalization
from keras.layers.advanced_activations import *
from keras.optimizers import RMSprop, Adam, SGD, Nadam
from keras.initializers import *
import ConfigParser
config = ConfigParser.RawConfigParser()
config.read('tmp/config.cfg')
except Exception as e:
if(logerrors):
logger.error(e)
else:
print e
#training parameters
ASK_FNAME = "rtf/eurusd002-ask.rtf"
BID_FNAME = "rtf/eurusd002-BID.rtf"
WINDOW=90
FORECAST=45
EMB_SIZE=10
STEP=1
TRAIN_TEST_PERCENTAGE=0.9
SAVE_NAME = "classification_model.hdf5"
ENABLE_CSV_OUTPUT = 1
NAME_CSV = "classification"
TRAINING = 1
TESTING = 0
NUMBER_EPOCHS = 10
TRADING_DAYS = 14
#execution parameters
try:
LOAD_NAME = config.get('model','LOAD_NAME')
except Exception as e:
if(logerrors):
logger.error(e)
def ternary_tensor(Time,aO,aH,aL,aC,aV,bO,bH,bL,bC,bV):
count=0
X,Y = [],[]
i=0
try:
count=count+1
try:
#ask open, ask high.. bid close, bid volume
ao = aO[i:i+WINDOW]
ah = aH[i:i+WINDOW]
al = aL[i:i+WINDOW]
ac = aC[i:i+WINDOW]
av = aV[i:i+WINDOW]
#zscore on time window interval
ao = (np.array(ao) - np.mean(ao)) / np.std(ao)
ah = (np.array(ah) - np.mean(ah)) / np.std(ah)
al = (np.array(al) - np.mean(al)) / np.std(al)
ac = (np.array(ac) - np.mean(ac)) / np.std(ac)
bo = bO[i:i+WINDOW]
bh = bH[i:i+WINDOW]
bl = bL[i:i+WINDOW]
bc = bC[i:i+WINDOW]
bv = bV[i:i+WINDOW]
#zscore on time window interval
bo = (np.array(bo) - np.mean(bo)) / np.std(bo)
bh = (np.array(bh) - np.mean(bh)) / np.std(bh)
bl = (np.array(bl) - np.mean(bl)) / np.std(bl)
bc = (np.array(bc) - np.mean(bc)) / np.std(bc)
x_i = np.column_stack((ao,ah,al,ac,av,bo,bh,bl,bc,bv))
except Exception as e:
print e
pass
except Exception as e:
print e
pass
return x_i
def format_data(df,ternary=1,binary=0):
Time = df.Datetime
aO = df.Open_x.tolist()
aH = df.High_x.tolist()
aL = df.Low_x.tolist()
aC = df.Close_x.tolist()
aV = df.Volume_x.tolist()
bO = df.Open_y.tolist()
bH = df.High_y.tolist()
bL = df.Low_y.tolist()
bC = df.Close_y.tolist()
bV = df.Volume_y.tolist()
#print(Time,aO,aH,aL,aC,aV,bO,bH,bL,bC)
if(ternary==1):
return ternary_tensor(Time,aO,aH,aL,aC,aV,bO,bH,bL,bC,bV)
elif(binary==1):
return binary_tensor(Time,aO,aH,aL,aC,aV,bO,bH,bL,bC,bV)
def spread(Y):
still = 0
for vec in Y:
if vec[2]==1:
still=still+1
spread =still*100/len(Y)
print spread,"%"
return spread
def shuffle_in_unison(a, b):
# courtsey http://stackoverflow.com/users/190280/josh-bleecher-snyder
# shuffling of training data
assert len(a) == len(b)
shuffled_a = np.empty(a.shape, dtype=a.dtype)
shuffled_b = np.empty(b.shape, dtype=b.dtype)
permutation = np.random.permutation(len(a))
for old_index, new_index in enumerate(permutation):
shuffled_a[new_index] = a[old_index]
shuffled_b[new_index] = b[old_index]
return shuffled_a, shuffled_b
def remove_nan_examples(data):
#some basic util functions
newX = []
for i in range(len(data)):
if np.isnan(data[i]).any() == False:
newX.append(data[i])
return newX
try:
#MODEL DEFINITION
print 'initializing model..'
model = Sequential()
model.add(Convolution1D(input_shape = (WINDOW, EMB_SIZE),
nb_filter=16,
filter_length=4,
border_mode='same'))
model.add(BatchNormalization())
model.add(LeakyReLU())
model.add(Dropout(0.5))
model.add(Convolution1D(nb_filter=8,
filter_length=4,
border_mode='same'))
model.add(BatchNormalization())
model.add(LeakyReLU())
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(64))
model.add(BatchNormalization())
model.add(LeakyReLU())
model.add(Dense(3))
model.add(Activation('softmax'))
opt = Nadam(lr=0.002)
reduce_lr = ReduceLROnPlateau(monitor='val_acc', factor=0.9, patience=30, min_lr=0.000001, verbose=1)
checkpointer = ModelCheckpoint(filepath=SAVE_NAME, verbose=1, save_best_only=True)
model.compile(optimizer=opt,
loss='categorical_crossentropy',
metrics=['accuracy'])
except Exception as e:
if(logerrors):
logger.error(e)
else:
print e
try:
import httplib
import json
from Tkinter import *
import datetime
import time
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.select import Select
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.keys import Keys
from pymouse import PyMouse
except Exception as e:
if(logerrors):
logger.error(e)
else:
print e
#app parameters
try:
CHROME_PATH = config.get('app','CHROME_PATH')
URL = config.get('app','URL')
USR = config.get('app','USR')
PSW = config.get('app','PSW')
except Exception as e:
logger.error(e)
class DukascopyBinary(object):
"""docstring for ClassName"""
def __init__(self, URL):
self.URL = URL
self.driver = webdriver.Chrome(CHROME_PATH)
self.driver.get(URL)
#datafeed values
self.minute=60
self.vec=[]
self.data=[]
def set_type(self, ASK=0, BID=1):
global USR
global PSW
time.sleep(10)
#logging in
self.login(USR,PSW)
time.sleep(15)
#1 minute time frame
self.driver.find_elements_by_css_selector(".S-T-U")[12].click()
time.sleep(2)
self.driver.find_element_by_id(":2c").click()
time.sleep(2)
#detached OHLC view *id :53 is variable attention
self.driver.find_elements_by_css_selector(".S-T-U")[52].click()
time.sleep(2)
self.driver.find_element_by_id(":53").click()
time.sleep(2)
#amount 100
self.driver.find_element_by_id("bp-amountfield-1074-inputEl").send_keys("00")
time.sleep(2)
#duration 45
self.driver.find_element_by_id("bp-numberfield-1078-inputEl").send_keys(Keys.BACKSPACE+"45")
time.sleep(2)
if(ASK):
self.type = "ASK"
self.driver.find_element_by_id(":1i").click()
time.sleep(2)
self.driver.find_element_by_id(":0").click()
time.sleep(2)
elif(BID):
self.type = "BID"
time.sleep(2)
for i in xrange(100):
self.driver.find_element_by_css_selector(".S-T-U.a-pb-gk-O").click()
time.sleep(0.02)
def set_mouse(self,m):
print 'put mouse in',self.type,' position.. (5s)'
time.sleep(5)
self.x,self.y=m.position()
print '-> position captured'
def login(self,USR,PSW):
self.driver.find_element_by_id("textfield-1020-inputEl").send_keys(USR)
self.driver.find_element_by_id("textfield-1021-inputEl").send_keys(PSW)
self.driver.find_element_by_id("button-1035-btnEl").click()
def call(self):
self.driver.find_element_by_class_name("call").click()
time.sleep(0.2)
self.driver.find_element_by_id("button-1014").click()
def put(self):
self.driver.find_element_by_class_name("put").click()
time.sleep(0.2)
self.driver.find_element_by_id("button-1014").click()
class App(object):
def __init__(self):
global model
model.load_weights(LOAD_NAME)
global URL
self.ask = DukascopyBinary(URL)
self.ask.set_type(ASK=1,BID=0)
self.bid = DukascopyBinary(URL)
self.bid.set_type(ASK=0,BID=1)
self.running = False
self.m = PyMouse()
self.data = []
self.temp = ""
self.datetime1,self.datetime2 = "",""
self.aOpen,self.aHigh,self.aLow,self.aClose,self.aVolume=0,0,0,0,0
self.bOpen,self.bHigh,self.bLow,self.bClose,self.bVolume=0,0,0,0,0
def mouse_config(self):
self.ask.set_mouse(self.m)
self.bid.set_mouse(self.m)
def start(self):
self.running = True
def stop(self):
self.running = False
def __tensor__(self):
print 'checking tensor'
if(len(self.data)>WINDOW):
df = pd.DataFrame(self.data[-WINDOW:])
df = df.rename(columns={ df.columns[0] : 'Datetime', df.columns[1] : 'Open_x', df.columns[2] : 'High_x', df.columns[3] : 'Low_x', df.columns[4]: 'Close_x', df.columns[5] : 'Volume_x', df.columns[6] : 'Open_y', df.columns[7] : 'High_y', df.columns[8] : 'Low_y', df.columns[9]: 'Close_y', df.columns[10] : 'Volume_y' })
self.X = format_data(df,1,0)
self.X = np.array(self.X)
return 1
else:
return 0
def __alg08__(self,pred,alfa):
if pred[0][0]-pred[0][2]>alfa:
self.ask.call()
print 'put executed'
elif pred[0][1]-pred[0][2]>alfa:
self.ask.put()
print 'call executed'
else:
print 'no signal'
def trading(self):
if self.running:
print 'RUNNING'
try:
now = app.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[0].get_attribute("innerHTML")+" "+app.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[1].get_attribute("innerHTML")
if now!=self.temp:
print 'NEW VALUE'
self.ask.driver.find_element_by_css_selector(".S-T-U.a-pb-gk-O").click()
self.bid.driver.find_element_by_css_selector(".S-T-U.a-pb-gk-O").click()
self.m.move(self.ask.x,self.ask.y)
self.m.move(self.bid.x,self.bid.y)
try:
print self.aOpen,self.aHigh,self.aLow,self.aClose,self.aVolume,self.bOpen,self.bHigh,self.bLow,self.bClose,self.bVolume
print 'NEW VALUE APPENDED',self.aClose
_time = datetime.datetime.strptime(app.datetime1+" "+app.datetime2,"%Y-%m-%d %H:%M:%S.%f")
self.data.append([_time,float(self.aOpen),float(self.aHigh),float(self.aLow),float(self.aClose),float(self.aVolume),float(self.bOpen),float(self.bHigh),float(self.bLow),float(self.bClose),float(self.bVolume)])
self.temp=now
if(self.__tensor__()):
pred =model.predict(np.reshape(self.X, (1,self.X.shape[0],self.X.shape[1])))
self.text.insert(1.0, pred)
self.__alg08__(pred, 0.495)
except Exception,e:
print 'inner loop exception:',str(e)
pass
print 'START SCANNING', self.aClose
self.datetime1 = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[0].get_attribute("innerHTML")
self.datetime2= self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[1].get_attribute("innerHTML")
self.aOpen = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[2].get_attribute("innerHTML")
self.aHigh = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[3].get_attribute("innerHTML")
self.aLow = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[4].get_attribute("innerHTML")
self.aClose = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[5].get_attribute("innerHTML")
self.aVolume = self.ask.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[6].get_attribute("innerHTML")
self.bOpen = self.bid.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[2].get_attribute("innerHTML")
self.bHigh = self.bid.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[3].get_attribute("innerHTML")
self.bLow = self.bid.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[4].get_attribute("innerHTML")
self.bClose = self.bid.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[5].get_attribute("innerHTML")
self.bVolume = self.bid.driver.find_elements_by_class_name("a-b-G-Ni-Oi")[6].get_attribute("innerHTML")
print 'FINISH SCANNING', self.aClose
except Exception,e:
print 'outer loop exception:',str(e)
pass
self.root.after(100, self.trading)
def run(self):
self.root = Tk()
self.root.title("v.0.4.1-framework")
self.root.geometry("300x100")
self.root2 = Tk()
self.text=Text(self.root2, width = 40, height=4, font=("Helvetica",32))
self.text.pack()
self.text.insert(1.0, "PREDICTION vals:")
app = Frame(self.root)
app.grid()
start = Button(app, text="Start", command=self.start)
stop = Button(app, text="Stop", command=self.stop)
start.grid(row=0, column=0, padx=(40,40), pady=(40,40))
stop.grid(row=0, column=1, padx=(40,40), pady=(40,40))
start.grid()
stop.grid()
self.root.after(1000, self.trading) # After 1 second, call scanning
self.root.mainloop()
self.root2.mainloop()
"""
to solve:
sync between ask and bid windows (when update time is not sync)
increase frequency of scanning?
ALWAYS double check with official DUKASCOPY data
"""
if(logerrors):
try:
app = App()
except Exception as e:
logger.error(e)
else:
app = App()
try:
app.mouse_config()
except Exception as e:
if(logerrors):
logger.error(e)
else:
print e
try:
app.run()
except Exception as e:
if(logerrors):
logger.error(e)
else:
print e