-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmodel.py
188 lines (167 loc) · 7.26 KB
/
model.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
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
#from sklearn.model_selection import train_test_split #废弃！！
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
#from sklearn.neural_network import BernoulliRBM
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn import metrics
from sklearn import linear_model
from sklearn.neural_network import MLPClassifier
from sklearn.metrics import confusion_matrix
from sklearn.decomposition import PCA
from imblearn.over_sampling import RandomOverSampler
from imblearn.under_sampling import ClusterCentroids,RandomUnderSampler
from imblearn.over_sampling import SMOTE, ADASYN
from collections import Counter
import math
h=0.2
# 读取训练数据
def readdata():
fake = pd.read_csv('fake.csv')
real = pd.read_csv('real.csv')
ALL=fake.append(real)
all=ALL.drop(['Unnamed: 0'],axis=1)
X=np.array(ALL)
#X = StandardScaler().fit_transform(X)
y=X[:,24]
#y=np.append(np.zeros(fake.shape[0]),np.ones(real.shape[0]))
X=X[:,1:-1]
return X,y
def allmodel():
classifiers = [
linear_model.LogisticRegression(C=1e5),#1
KNeighborsClassifier(5),#2
SVC(kernel="linear", C=0.025),#3
SVC(gamma='auto', C=1),#4
DecisionTreeClassifier(max_depth=5),#5
RandomForestClassifier(max_depth=5, n_estimators=10, max_features=1),#6
AdaBoostClassifier(),#7
GaussianNB(),#8
MLPClassifier(solver='lbfgs', alpha=1e-5,hidden_layer_sizes=(5, 2), random_state=1),#9
]
names = ['LogisticRegression',#1
"Nearest Neighbors", #2
"Linear SVM", #3
"RBF SVM",#4
"Decision Tree",#5
"Random Forest", #6
"AdaBoost",#7
"Naive Bayes",#8
'MLPClassifier',#9
]
index=['origional',
"overSampler",
'underSampler',
'smotesampler',
'adasynsampler'
]
return classifiers,names,index
########################
#print('%-15s %-15s %-15s'%('name','score','roc_auc','wrong'))
def runmodel(input_x,input_y,index_name):
X_train, X_test, y_train, y_test = train_test_split(input_x,input_y, test_size=.5, random_state=1)
print("Method ACC AUC RECALL")
for name, clf in zip(names, classifiers):
# ax = plt.subplot(len(datasets), len(classifiers) + 1, i)
clf.fit(X_train, y_train)
score = clf.score(X_test, y_test)
#confusion=confusion_matrix(y_test, clf.predict(X_test))
#
re=metrics.recall_score(y_test, clf.predict(X_test), average=None)[0]
#metrics.f1_score(y_test, clf.predict(X_test))
fpr, tpr, thresholds = metrics.roc_curve(y_test, clf.predict(X_test))
roc_auc = metrics.auc(fpr, tpr)
print('%-20s, %f, %f, %f '%(name,score,roc_auc,re))
pre_score[name][index_name]=np.array([score,roc_auc,re])
# print('%-15s %-15s %-15s %-15s '%(name,score,roc_auc,sum(abs(clf.predict(X_test)-y_test))))
#,confusion)
#pred =
#print(metrics.classification_report(y_test, clf.predict(X_test)))
#######################
def runmodel_tra(input_x,input_y,index_name):
X_train, X_test, y_train, y_test = train_test_split(input_x,input_y, test_size=.5, random_state=1)
X_test,y_test=X_train,y_train
print("Method ACC AUC RECALL")
for name, clf in zip(names, classifiers):
# ax = plt.subplot(len(datasets), len(classifiers) + 1, i)
clf.fit(X_train, y_train)
score = clf.score(X_test, y_test)
#confusion=confusion_matrix(y_test, clf.predict(X_test))
#
re=metrics.recall_score(y_test, clf.predict(X_test), average=None)[0]
#metrics.f1_score(y_test, clf.predict(X_test))
fpr, tpr, thresholds = metrics.roc_curve(y_test, clf.predict(X_test))
roc_auc = metrics.auc(fpr, tpr)
print('%-20s, %f, %f, %f '%(name,score,roc_auc,re))
tra_score[name][index_name]=np.array([score,roc_auc,re])
#####################
def plot2(X,y,title):
pca = PCA(n_components=2)
newData=pca.fit_transform(X)
cValue=[]
for i in range(y.size):
if y[i]==0:
cValue.append('r')
else:
cValue.append('b')
plt.scatter(newData[:,0], newData[:,1] , c=cValue, marker='o') #cmap=plt.cm.Paired
title=title+str(sorted(Counter(y.astype(int)).items()))
plt.title(title)
plt.show()
##################
def overSampler(X,y):
ros = RandomOverSampler(random_state=0)
X_resampled, y_resampled = ros.fit_sample(X, y)
return X_resampled,y_resampled
##################
def underSampler(X,y):
cc = RandomUnderSampler(random_state=0)
X_resampled, y_resampled = cc.fit_sample(X, y)
return X_resampled,y_resampled
######################
def smotesampler(X,y):
X_resampled_smote, y_resampled_smote = SMOTE().fit_sample(X, y)
return X_resampled_smote, y_resampled_smote
def adasynsampler(X,y):
X_resampled_adasyn, y_resampled_adasyn = ADASYN().fit_sample(X, y)
return X_resampled_adasyn, y_resampled_adasyn
#######################
#sorted(Counter(y).items())
if __name__ == '__main__':
classifiers,names,index=allmodel()
X,y=readdata()
X_resampled_over,y_resampled_over=overSampler(X,y)
X_resampled_under,y_resampled_under=underSampler(X,y)
X_resampled_smote, y_resampled_smote=smotesampler(X,y)
X_resampled_adasyn, y_resampled_adasyn=adasynsampler(X,y)
plot2(X,y,"All DATA(red=fake,blue=real)")
plot2(X_resampled_over,y_resampled_over,"overSampler(red=fake,blue=real)")
plot2(X_resampled_under,y_resampled_under,"underSampler(red=fake,blue=real)")
plot2(X_resampled_smote, y_resampled_smote,"smote(red=fake,blue=real)")
plot2(X_resampled_adasyn, y_resampled_adasyn,"adasyn(red=fake,blue=real)")
pre_score=pd.DataFrame(index=index,columns=names)
tra_score=pd.DataFrame(index=index,columns=names)
runmodel(X,y,index[0])
runmodel(X_resampled_over,y_resampled_over,index[1])
runmodel(X_resampled_under,y_resampled_under,index[2])
runmodel(X_resampled_smote, y_resampled_smote,index[3])
runmodel(X_resampled_adasyn, y_resampled_adasyn,index[4])
runmodel_tra(X,y,index[0])
runmodel_tra(X_resampled_over,y_resampled_over,index[1])
runmodel_tra(X_resampled_under,y_resampled_under,index[2])
runmodel_tra(X_resampled_smote, y_resampled_smote,index[3])
runmodel_tra(X_resampled_adasyn, y_resampled_adasyn,index[4])
for name, clf in zip(names, classifiers):
print('\n%-20s, %-15s, %s, %s '%(name,'pre_auc','tra_auc','alpha'))
for index_name in index:
pre_auc=pre_score[name][index_name][1]
tra_auc=tra_score[name][index_name][1]
myscore=-math.log(pre_auc/tra_auc)
print('%-20s, %-15s, %f, %f, %f '%(name,index_name,pre_auc,tra_auc,myscore))