Machine-Learning/Regression.py at master · Yan1015/Machine-Learning · GitHub

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

# coding: utf-8


import pandas as pd
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
from sklearn.linear_model import SGDClassifier
from sklearn.linear_model import Ridge
from sklearn.linear_model import Lasso


data=pd.read_csv('data.csv')
del data['mid_weighted_average_price']


def LinearModel(model,data,interval):#using LinearRegression and LogisticRegression
    clf=model
    train=data.iloc[0:int(data.shape[0]*0.7)]#split data to 70% train and 30% test
    test=data.iloc[int((data.shape[0])*0.7+1):-1]
    test.index=range(int(data.shape[0]*0.3-1))
    y_train=train['best_mid_price'].iloc[interval:-1]#set 'best_mid_price' after a certain interval as the target values
    y_test=test['best_mid_price'].iloc[interval:-1]
    clf.fit(train.iloc[0:-interval-1,2:10],y_train)
    predict1=clf.predict(train.iloc[0:-interval-1,2:10])#for calculating the bias
    predict2=clf.predict(test.iloc[0:-interval-1,2:10])#predict the target values in test data


    xx=np.zeros(y_train.shape[0])#collecting the bias for each record
    for i in range(y_train.shape[0]):
        xx[i]=predict1[i]-y_train.iloc[i]


    xx2=np.zeros(y_test.shape[0])
    yy=np.zeros(y_test.shape[0])
    for i in range(y_test.shape[0]):#determine the prediction
        xx2[i]=predict2[i]-y_test.iloc[i]
        if xx2[i]<-0.01:
            yy[i]=-1
        elif xx2[i]>0.01:
            yy[i]=1
        else:
            yy[i]=0
    s=0
    index=int(data.shape[0]*0.7+1)
    for i in range(len(yy)):
        if int(yy[i])==data['output_30'].iloc[index+i]:
            s+=1

    return yy,s/len(yy)


def LinearClassifier(data,interval,column):# using SGDClassifier
    clf=SGDClassifier()
    train=data.iloc[0:int(data.shape[0]*0.7)]
    test=data.iloc[int((data.shape[0])*0.7+1):-1]
    test.index=range(int(data.shape[0]*0.3-1))
    y_train=train.iloc[interval:-1,10+column]
    y_test=test.iloc[interval:-1,10+column]
    clf.fit(train.iloc[0:-interval-1,2:10],y_train)
    predict=clf.predict(test.iloc[0:-interval-1,2:10])
    score=clf.score(test.iloc[0:-interval-1,2:10],y_test)
    return predict,score


def ret_cal(seq,data,interval):
    ret=0
    for i in range(seq.shape[0]):
        if seq[i]==-1:
            ret+=(data.iloc[i]-data.iloc[i+interval])/data.iloc[i]
        elif seq[i]==1:
            ret+=(data.iloc[i+interval]-data.iloc[i])/data.iloc[i]
    return ret


output=pd.DataFrame([])
score=np.zeros(18)
s=0
for i in ['1','5','10','50','100','300']:
    yy,score[s]=LinearModel(LinearRegression(),data,int(i))#yy is an array of output, score is the accuracy of prediction
    yy=yy[0:17505]
    output[str(s)]=yy
    yy,score[s+1]=LinearModel(LogisticRegression(max_iter=10),data,int(i))#max_iter cannot be large or it will cost minutes to fit
    yy=yy[0:17505]
    output[str(s+1)]=yy
    yy,score[s+2]=LinearClassifier(data,int(i),int(s/3+1))
    yy=yy[0:17505]
    output[str(s+2)]=yy
    s=s+3
print(score)
output.columns=['lin_0.1','log_0.1','sgd_0.1','lin_0.5','log_0.5','sgd_0.5',
               'lin_1','log_1','sgd_1','lin_5','log_5','sgd_5',
                'lin_10','log_10','sgd_10','lin_30','log_30','sgd_30'
               ]


ret_cal(output.iloc[:,0],data["best_mid_price"].iloc[41559:-1],300)


t=0
for i in ['1','5','10','50','100','300','1','5','10','50','100','300','1','5','10','50','100','300']:
    ret[t]=ret_cal(output.iloc[:,t],data["best_mid_price"].iloc[41559:-1],int(i))
    t+=1


output.loc['return']=ret
output.loc['score']=score


output.to_csv('C:/Users/Hao Li/Desktop/2018 spring/9733/project/lin_log_sgd.csv')