-
Notifications
You must be signed in to change notification settings - Fork 0
/
tmp.py
127 lines (115 loc) · 5.15 KB
/
tmp.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
# coding=utf-8
import numpy as np
import os
# import pandas as pd
import warnings
from itertools import combinations
# from getRep import getReptf
from functools import partial
from datetime import datetime
# import matplotlib.pyplot as plt
from sys import argv
# import face
import xlwt
from sklearn import metrics
# model_name = face.facenet_model_checkpoint.split('/')[-1]
model_name = "tmp"
if __name__ == '__main__':
now = datetime.strftime(datetime.now(), '%Y-%m-%d-%H_%M_%S')
thresholds = np.arange(0.01, 1.8, 0.01)
P_valid_force = 45000 # sample number need to change
N_valid_force = 1000000 #100w
best_precison = 0
best_accuracy = 0
best_threshold_precison=None
best_threshold_acc=None
myRound = partial(round,ndigits=4) #偏函数,2018.08.06，by xjxf
test_file =argv[1]
savedir = os.path.join('./saveResults/',now)
if not os.path.exists(savedir):
os.makedirs(os.path.abspath(savedir))
f1 = open(os.path.join(savedir,'TestReport%s.txt'% now),'wb')
f2 = open(os.path.join(savedir,'TestReport%s_backup.txt'% now),'wb')
# 创建excel文件指针,2018.08.07,by xjxf _start
path_excel=os.path.join(savedir,'oush_distance%s_data.xls'% now)
xls_3=xlwt.Workbook()
sheet_1=xls_3.add_sheet('sheet_1',cell_overwrite_ok=True)
row_count=0
# 创建excel文件指针,2018.08.07,by xjxf _end
f1.write('-----测试日期:{}, 测试模型：{}, 测试数据:{},距离指标:欧式距离-----\n'.format(now,model_name,test_file))
f2.write('-----测试日期:{}, 测试模型：{}, 测试数据:{},距离指标:欧式距离-----\n'.format(now,model_name,test_file))
print '-----测试日期:{}, 测试模型：{}, 测试数据:{},,距离指标:欧式距离------\n'.format(now,model_name,test_file)
baseData = baseImageRep_tf(test_file) # 提取基础集的特征点和标签
pairs_list_all = list(combinations(baseData, 2))
np.random.shuffle(pairs_list_all)
P_num = 0
N_num = 0
pair_list =[]
for pair in pairs_list_all:
if pair[0][0]==pair[1][0] and P_num<P_valid_force:
pair_list.append(pair)
P_num+=1
elif pair[0][0]!=pair[1][0] and N_num<N_valid_force:
pair_list.append(pair)
N_num+=1
elif P_num==P_valid_force and N_num== N_valid_force:
break
s_ = '**正测试单元数量:{}, 负测试单元数量:{}**\n'.format(P_num,N_num)
print s_
f1.write(s_)
f2.write(s_)
print("P_valid_force is " + str(P_valid_force))
print("P_num in code is " + str(P_num))
print("N_valid_force is " + str(N_valid_force))
print("N_num in code is " + str(N_num))
# # fig = plt.figure(figsize=(30,50))
# # dist_P = np.array(distList)[actual_issameList].tolist()
# # dist_N = np.array(distList)[np.logical_not(actual_issameList)].tolist()
# # plt.scatter(np.arange(len(dist_P)),dist_P,s=1000,marker='*',label='Positive')
# # plt.title(now,fontsize=70)
# # plt.scatter(np.arange(len(dist_N)), dist_N, s=1000, marker='.',label='Negative')
# # plt.yticks(fontsize=50)
# # plt.legend(loc='upper right',fontsize=50)
# # plt.savefig(os.path.join(savedir,'%s.png'% now))
# # plt.show()
# f2.write('threshold,recall, precision, acc, tnr, fpr,fnr\n')
# auc={"fpr_acu":[],"tpr_auc":[]}
# for threshold in thresholds:
# print '---------threshold:',round(threshold,2),'---------------\n'
# f1.write('---------threshold:'+str(round(threshold,2))+'---------------\n')
#
# recall, precision, acc, tnr, fpr,fnr = map(myRound,calculate_accuracy(threshold, distList, actual_issameList))
# auc['fpr_acu'].append(fpr)
# auc['tpr_auc'].append(recall)
# if precision>best_precison:
# best_precison = precision
# best_threshold_precison = threshold
#
# if acc>best_accuracy:
# best_accuracy=acc
# best_threshold_acc = threshold
#
# ss = 'recall召回率:'+str(recall)+'\t'+'precison精准率:'+str(precision)+'\t'+' accuracy正确率:'+str(acc)+'\n'
# sss = 'TNR真负率:'+str(tnr)+'\t'+'FPR假正率:'+str(fpr)+'\t'+'FNR假负率:'+str(fnr)+'\t\n\n'
# print ss,sss
# f1.write(ss)
# f1.write(sss)
# f2.write('%f,%f,%f,%f,%f,%f,%f\n'%(threshold,recall, precision, acc, tnr, fpr,fnr))
# #写excel文件,2018.08.07, by xjxf _start
# list_data=[threshold,tnr, fpr,fnr,recall, precision, acc]
# for i_1 in range(len(list_data)):
# sheet_1.write(row_count,i_1,list_data[i_1])
# row_count=row_count+1
# xls_3.save(path_excel)
#
# # 写excel文件,2018.08.07, by xjxf _end
#
# foo = '\n\n*********conclusion********\n'+'best_precison:{} best_threshold_precision:{}\nbest_accuracy:{} best_threshold_acc:{} '.format(best_precison, best_threshold_precison,best_accuracy,best_threshold_acc)
# print foo
# f1.write(foo)
#
# print("model_name",model_name)
# print("data path:",test_file)
# print("excel path:",path_excel)
# auc_result = metrics.auc(auc['fpr_acu'], auc['tpr_auc'])
# print('Area Under Curve (AUC): %1.3f' % auc_result)