-
Notifications
You must be signed in to change notification settings - Fork 90
/
Copy path3D_predict.py
205 lines (171 loc) · 7.76 KB
/
3D_predict.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
#coding = utf-8
#Fire / 2018.6.30
import requests
from bs4 import BeautifulSoup as bs
import re
import time
import keras
import pandas as pd
import numpy as np
from keras.models import Sequential
from keras.layers.recurrent import LSTM
from keras.layers.wrappers import TimeDistributed
from keras.layers.core import Dense, RepeatVector
from keras.layers import Dropout
#玄学参数 随机种子
win_number = 2018
from numpy.random import seed
seed(win_number)
from tensorflow import set_random_seed
set_random_seed(win_number)
class predict3D(object):
"""docstring for predict3D"""
def __init__(self, series='-1', histotyFile='history3D.txt'):
self.series = series
self.histotyFile = histotyFile
self.data = None #最新的数据在最后 str格式
self.latest = -1 #str
self.model1 = None
self.model2 = None
self.model3 = None
self.ball1_train = None
self.ball1_label = None
self.ball2_train = None
self.ball2_label = None
self.ball3_train = None
self.ball3_label = None
self.seq_N = 10
self.batch_size = 1024
self.epochs = 100
self.validation_split = 0.01
self.hidden_size = 64
def readHistory(self):
with open(self.histotyFile,'r') as f:
raw_data = f.readlines()
self.latest = raw_data[0].split(':')[0]
#print(int(''.join(raw_data[0].split(':')[1].split(' ')[:-1])))
data = []
for line in raw_data:
data.append(''.join(line.split(':')[1].split(' ')[:-1]))
self.data = data[::-1] #最新的数据在最后
print('Finish read local data: ',len(self.data))
def getLatestNumber(self):
url = "https://datachart.500.com/sd/history/inc/history.php?limit=18173&start=2000000&end=2018173"
headers = {
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
'Accept-Encoding': 'gzip, deflate, sdch',
'Accept-Language': 'zh-CN,zh;q=0.8,en;q=0.6',
'Cache-Control': 'max-age=0',
'Connection': 'keep-alive',
'Host': 'datachart.500.com',
'Upgrade-Insecure-Requests': '1',
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.102 Safari/537.36'
}
r = requests.get(url=url, headers=headers)
r.encoding = 'gb2312'
html = r.text
# 解析html
soup = bs(html, 'lxml')
ul = soup.find_all(name='tr',attrs={'class': 't_tr1'})
latest = str(ul[0].find_next(name='td'))[4:11]
if latest == self.latest:
print("本地已经是最新数据")
else:
print("本地数据数: ", len(self.data))
dist = int(latest) - int(self.latest)
newdata = []
for i in range(dist):
newdata.append(''.join(str(ul[i].find_next(name='td',attrs={'class': 'cfont2'})).split('>')[1][:5].split(' ')))
newdata = newdata[::-1]
self.data = self.data + newdata
print("添加后本地数据数: ", len(self.data))
def buildModel(self, input_size = 10, seq_len = None, hidden_size=None):
if not seq_len:
seq_len = self.seq_N
if not hidden_size:
hidden_size = self.hidden_size
model = Sequential()
model.add(LSTM(hidden_size, input_shape=( seq_len,input_size)))
model.add(Dropout(0.2))
model.add(Dense(input_size, activation='softmax'))
model.compile(loss='categorical_crossentropy', optimizer='adam')
#print(model.summary())
return model
def makeModelData(self):
self.readHistory()
self.getLatestNumber()
seq_N = self.seq_N
ball1_train = []
ball1_label = []
ball2_train = []
ball2_label = []
ball3_train = []
ball3_label = []
for i in range(seq_N,len(self.data)):
data_tmp = self.data[i-seq_N:i][::-1] ## 小trick。逆序
x_tmp = [int(x[0]) for x in data_tmp]
x_tmp = keras.utils.np_utils.to_categorical(x_tmp, 10)
ball1_train.append(x_tmp)
y_tmp = int(self.data[i][0])
y_tmp = keras.utils.np_utils.to_categorical(y_tmp, 10)
ball1_label.append(y_tmp)
x_tmp = [int(x[1]) for x in data_tmp]
x_tmp = keras.utils.np_utils.to_categorical(x_tmp, 10)
ball2_train.append(x_tmp)
y_tmp = int(self.data[i][1])
y_tmp = keras.utils.np_utils.to_categorical(y_tmp, 10)
ball2_label.append(y_tmp)
x_tmp = [int(x[2]) for x in data_tmp]
x_tmp = keras.utils.np_utils.to_categorical(x_tmp, 10)
ball3_train.append(x_tmp)
y_tmp = int(self.data[i][2])
y_tmp = keras.utils.np_utils.to_categorical(y_tmp, 10)
ball3_label.append(y_tmp)
ball1_train = np.array(ball1_train)
ball1_label = np.array(ball1_label)
ball1_label = np.reshape(ball1_label,(ball1_label.shape[0],ball1_label.shape[2]))
ball2_train = np.array(ball2_train)
ball2_label = np.array(ball2_label)
ball2_label = np.reshape(ball2_label,(ball2_label.shape[0],ball2_label.shape[2]))
ball3_train = np.array(ball3_train)
ball3_label = np.array(ball3_label)
ball3_label = np.reshape(ball3_label,(ball3_label.shape[0],ball3_label.shape[2]))
self.ball1_train = ball1_train
self.ball1_label = ball1_label
self.ball2_train = ball2_train
self.ball2_label = ball2_label
self.ball3_train = ball3_train
self.ball3_label = ball3_label
print("Make data finished.And ball 1 data lenth: ",ball1_train.shape, ball1_label.shape)
def train(self):
self.makeModelData()
self.model1 = self.buildModel()
self.model1.fit(self.ball1_train, self.ball1_label,batch_size=self.batch_size,epochs=self.epochs,validation_split=self.validation_split)
self.model2 = self.buildModel()
self.model2.fit(self.ball2_train, self.ball2_label,batch_size=self.batch_size,epochs=self.epochs,validation_split=self.validation_split)
self.model3 = self.buildModel()
self.model3.fit(self.ball3_train, self.ball3_label,batch_size=self.batch_size,epochs=self.epochs,validation_split=self.validation_split)
print('Finish train.')
def predict(self):
data_tmp = self.data[-self.seq_N:][::-1] ## 小trick。逆序
true_last1 = [int(x[0]) for x in data_tmp]
true_last1 = keras.utils.np_utils.to_categorical(true_last1, 10)
test_x1 = np.reshape(true_last1, (1, self.seq_N, 10))
test_y1 = self.model1.predict(test_x1)
test_y1 = np.argmax(test_y1)
true_last2 = [int(x[1]) for x in data_tmp]
true_last2 = keras.utils.np_utils.to_categorical(true_last2, 10)
test_x2 = np.reshape(true_last2, (1, self.seq_N, 10))
test_y2 = self.model2.predict(test_x2)
test_y2 = np.argmax(test_y2)
true_last3 = [int(x[2]) for x in data_tmp]
true_last3 = keras.utils.np_utils.to_categorical(true_last3, 10)
test_x3 = np.reshape(true_last3, (1, self.seq_N, 10))
test_y3 = self.model3.predict(test_x3)
test_y3 = np.argmax(test_y3)
print("上期期数：",self.latest, "上期号码：", self.data[-1])
pre = 0
print("本期预测号码：", test_y1, test_y2, test_y3)
model = predict3D()
model.train()
model.predict()