-
Notifications
You must be signed in to change notification settings - Fork 0
/
classification_cost.py
190 lines (158 loc) · 8.08 KB
/
classification_cost.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
import pandas as pd
import numpy as np
import csv
import os
from sklearn.linear_model import LogisticRegression
from sklearn.tree import DecisionTreeClassifier
from scripts.classification_utils import prep_data, add_values_in_dict
from scripts.evaluation_utils import evaluating_model
def save_dict_in_csv(results_dict, fieldnames, name_csv):
"""
Save dictionary as csv.
# Reference: https://stackoverflow.com/questions/53013274/writing-data-to-csv-from-dictionaries-with-multiple-values-per-key
Args:
- results_dict <dict> results that should be saved
- fieldnames <list>: the col names for csv
- name_csv <str>: name and path of csv file
"""
# Example for fieldnames:
# overall_fieldnames = ['Run', 'Acc', 'F1micro/F1w/F1bsr', 'SelectionRate', 'TNR rate', 'TPR rate', 'FNER', 'FPER', 'DIB/DIW', 'DP Diff', 'EO Diff']
# byrace_fieldnames = ['Run', 'Acc', 'F1micro/F1w/F1bsr', 'SelectionRate', 'TNR rate', 'TPR rate', 'FNER', 'FPER', 'DIB/DIW']
#mode = 'a' if os.path.exists(name_csv) else 'w+'
#print(mode)
#if mode == 'w+':
# os.makedirs(name_csv, exist_ok=True)
# the dictionary needs to be formatted like: {'Run1': [acc, f1, ...], 'Run2': [acc, f1, ...]}
with open(name_csv, mode='w+') as csv_file:
writer = csv.writer((csv_file))
writer.writerow(fieldnames)
for run in results_dict.items():
##print(run)
##print([row[0]])
##print(row[1])
row = list(run)
row = [row[0]] + row[1]
writer.writerow(row)
csv_file.close()
"""
PARAMETER SETTING
"""
data_path = 'data/synthetic_datasets/Demo-0-Lab-0.csv'# path to the dataset csv-file
fp_weight = 6
fn_weight = 5
results_path = 'fico-results/cost-test/' # directory to save the results
balanced = True
random_bool = True
weight_idx = 1 # weight index for samples (1 in our runs)
testset_size = 0.3 # proportion of testset samples in the dataset (e.g. 0.3)
test_set_variant = 0 # 0= default (testset like trainset), 1= balanced testset, 2= original,true FICO distribution
test_set_bound = 30000 # absolute upper bound for test_set size
constraint_str = 'Cost-'
di_means = [75,-150] # means for delayed impact distributions (rewardTP,penaltyFP)
di_stds = [15,15] # standard deviations for delayed impact distributions (rewardTP,penaltyFP)
save = True # indicator if the results should be saved
models = {'Decision Tree': 'dt', 'Logistic Regression': 'lgr'}
model_name = models['Logistic Regression']
if balanced:
run_key = f'{model_name}cost-balance'
results_path_full = results_path + model_name + f'/cost-balance/'
else:
run_key = f'{model_name}cost-fp{fp_weight}-fn{fn_weight}'
results_path_full = results_path+model_name+f'/cost-fp{fp_weight}-fn{fn_weight}/'
overall_results_dict = {}
black_results_dict = {}
white_results_dict = {}
combined_results_dict = {}
all_types = []
all_scores = []
scores_names = []
data = pd.read_csv(data_path)
data[['score', 'race']] = data[['score', 'race']].astype(int)
x = data[['score', 'race']].values
y = data['repay_indices'].values
X_train, X_test, y_train, y_test, race_train, race_test, sample_weight_train, sample_weight_test = prep_data(data, testset_size,test_set_variant,test_set_bound, weight_idx)
X_test_b = []
X_test_w = []
y_test_b = []
y_test_w = []
for index in range(len(X_test)):
if race_test[index] == 0: # black
X_test_b.append(X_test[index][0])
y_test_b.append(y_test[index])
elif race_test[index] == 1: # white
X_test_w.append(X_test[index][0])
y_test_w.append(y_test[index])
# NOTE: I DIDN'T INCLUDE THE SAVING OF SCORES AND TYPES TO A LIST
results_path_full += f'{model_name}/'
print(results_path_full)
if not os.path.exists(results_path_full):
os.makedirs(results_path_full, exist_ok=True)
"""
MODEL TRAINING
"""
print('The classifier trained below is: ', model_name)
if model_name == 'lgr':
if random_bool:
if not balanced:
classifier = LogisticRegression(class_weight={0:fp_weight, 1:fn_weight}, random_state = 0)
else:
classifier = LogisticRegression(class_weight='balanced', random_state = 0)
else:
if not balanced:
classifier = LogisticRegression(class_weight={0:fp_weight, 1:fn_weight})
else:
classifier = LogisticRegression(class_weight='balanced')
elif model_name == 'dt':
if random_bool:
if not balanced:
classifier = DecisionTreeClassifier(class_weight={0:fp_weight, 1:fn_weight}, random_state = 0)
else:
classifier = DecisionTreeClassifier(class_weight='balanced', random_state = 0)
else:
if not balanced:
classifier = DecisionTreeClassifier(class_weight={0:fp_weight, 1:fn_weight})
else:
classifier = DecisionTreeClassifier(class_weight='balanced')
else:
print('error: input an available model, lgr or dt')
# Resource: https://fraud-detection-handbook.github.io/fraud-detection-handbook/Chapter_6_ImbalancedLearning/CostSensitive.html
# {0:c10 (FP), 1:c01 (FN)}: The misclassification costs are explicitly set for the two classes by means of a dictionary.
# Conf matrix: [c00, c01(FN)]
# [c10(FP), c11]
# Reference: https://www.datacamp.com/community/tutorials/decision-tree-classification-python
np.random.seed(0)
# Train the classifier:
model = classifier.fit(X_train,y_train, sample_weight_train)
# Make predictions with the classifier:
y_predict = model.predict(X_test)
# Scores on test set
test_scores = model.predict_proba(X_test)[:, 1]
# results_overall = accuracy, cs_matrix, f1_micro, f1_weighted, f1_binary, round(sr * 100, 2), tnr, tpr, fner, fper,
# di_B, di_W, round(dp_diff * 100, 2), round(eod_diff * 100, 2), round(eoo_dif * 100, 2),
# round(fpr_dif * 100, 2), round(er_dif * 100, 2)]
# results_0 = [accuracy_0, cs_m_0, f1_m_0, f1_w_0, f1_b_0, sr_0, tnr_0, tpr_0, fner_0, fper_0, round(di_0, 2)]
# results_1 = [accuracy_1, cs_m_1, f1_m_1, f1_w_1, f1_b_1, sr_1, tnr_1, tpr_1, fner_1, fper_1, round(di_1, 2)]
results_overall, results_black, results_white = evaluating_model(constraint_str,X_test,y_test, y_predict, di_means,di_stds, sample_weight_test,race_test)
combined_results = [results_overall[3], results_overall[0], results_overall[5], results_overall[6],
results_overall[7], results_overall[8], results_overall[9], results_overall[10],
results_overall[11], results_black[6], results_black[7], results_black[8], results_black[9],
results_white[6], results_white[7], results_white[8], results_white[9]]
"""
SAVING RESULTS
"""
overall_results_dict = add_values_in_dict(overall_results_dict, run_key, results_overall)
black_results_dict = add_values_in_dict(black_results_dict, run_key, results_black)
white_results_dict = add_values_in_dict(white_results_dict, run_key, results_white)
combined_results_dict = add_values_in_dict(combined_results_dict, run_key, combined_results)
# To use below!!
if save == True:
overall_fieldnames = ['Run', 'Acc', 'ConfMatrix','F1micro', 'F1weighted','F1binary', 'SelectionRate', 'TNR rate', 'TPR rate', 'FNER', 'FPER', 'DIB','DIW', 'DP Diff', 'EO Diff', 'TPR Diff', 'FPR Diff', 'ER Diff']
byrace_fieldnames = ['Run', 'Acc', 'ConfMatrix','F1micro', 'F1weighted','F1binary', 'SelectionRate', 'TNR rate', 'TPR rate', 'FNER', 'FPER', 'DI']
combined_fieldnames = ['Run', 'F1_weighted', 'Acc', 'SelectionRate', 'TNR', 'TPR', 'FNER', 'FPER',
'Black Impact', 'White Impact', 'TNR_B', 'TPR_B', 'FNER_B', 'FPER_B', 'TNR_W', 'TPR_W',
'FNER_W', 'FPER_W']
save_dict_in_csv(overall_results_dict, overall_fieldnames, results_path_full+model_name+'_overall_results.csv')
save_dict_in_csv(black_results_dict, byrace_fieldnames, results_path_full+model_name+'_black_results.csv')
save_dict_in_csv(white_results_dict, byrace_fieldnames, results_path_full+model_name+'_white_results.csv')
save_dict_in_csv(combined_results_dict, combined_fieldnames,
results_path_full + model_name + '_combined_results.csv')