-
Notifications
You must be signed in to change notification settings - Fork 1
/
auc_decompose.py
147 lines (125 loc) · 5.83 KB
/
auc_decompose.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
import pandas as pd
import os
from support.acc_funs import auc_decomp
dir_base = os.getcwd()
dir_output =os.path.join(dir_base, '..', 'output')
dir_figures = os.path.join(dir_base, '..', 'figures')
fn_Y = 'y_agg.csv'
dat_Y = pd.read_csv(os.path.join(dir_output, fn_Y))
cn_Y = list(dat_Y.columns[25:37])
############### LOGIT
# AGG
read_file_1 = 'best_agg_logit.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
agg_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': agg_auc_decomp.tt, 'auc': agg_auc_decomp.auc,
'den': agg_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
agg_model_auc = pd.concat(result_list).reset_index(drop=True)
agg_model_auc.to_csv(os.path.join(dir_output, 'logit_agg_model_auc_decomposed.csv'), index=False)
###############################################
# decompose auc and save
read_file_1 = 'best_sub_logit.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
sub_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': sub_auc_decomp.tt, 'auc': sub_auc_decomp.auc,
'den': sub_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
sub_model_auc = pd.concat(result_list).reset_index(drop=True)
sub_model_auc.to_csv(os.path.join(dir_output, 'logit_sub_model_auc_decomposed.csv'), index=False)
############### random forest
###############################################
# decompose auc and save
read_file_1 = 'best_agg_rf.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
agg_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': agg_auc_decomp.tt, 'auc': agg_auc_decomp.auc,
'den': agg_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
agg_model_auc = pd.concat(result_list).reset_index(drop=True)
agg_model_auc.to_csv(os.path.join(dir_output, 'rf_agg_model_auc_decomposed.csv'), index=False)
###############################################
# decompose auc and save
read_file_1 = 'best_sub_rf.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
sub_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': sub_auc_decomp.tt, 'auc': sub_auc_decomp.auc,
'den': sub_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
sub_model_auc = pd.concat(result_list).reset_index(drop=True)
sub_model_auc.to_csv(os.path.join(dir_output, 'rf_sub_model_auc_decomposed.csv'), index=False)
############### xgboost
###############################################
# decompose auc and save
read_file_1 = 'best_agg_xgb.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
agg_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': agg_auc_decomp.tt, 'auc': agg_auc_decomp.auc,
'den': agg_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
agg_model_auc = pd.concat(result_list).reset_index(drop=True)
agg_model_auc.to_csv(os.path.join(dir_output, 'xgb_agg_model_auc_decomposed.csv'), index=False)
###############################################
# decompose auc and save
read_file_1 = 'best_sub_xgb.csv'
res_y_all = pd.read_csv(os.path.join(dir_output, read_file_1))
res_y_all = res_y_all.dropna().reset_index(drop=True)
result_list = []
for i in cn_Y:
print(i)
temp_list = []
sub_res = res_y_all[res_y_all['outcome']==i].reset_index(drop=True)
y_values = sub_res.y.values
pred_values = sub_res.preds.values
group_values = sub_res.cpt.values
sub_auc_decomp = auc_decomp(y=y_values, score=pred_values, group=group_values, rand=False)
temp_list.append(pd.DataFrame({'tt': sub_auc_decomp.tt, 'auc': sub_auc_decomp.auc,
'den': sub_auc_decomp.den}))
result_list.append(pd.concat(temp_list).assign(outcome=i))
sub_model_auc = pd.concat(result_list).reset_index(drop=True)
sub_model_auc.to_csv(os.path.join(dir_output, 'xgb_sub_model_auc_decomposed.csv'), index=False)