-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathinitial_model.py
155 lines (139 loc) · 8.5 KB
/
initial_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
import numpy as np
import timeit
import datetime
import argparse
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense
from keras import backend as K
import read_data
import group_fairness
import individual_fairness
import os
import tensorflow as tf
parser = argparse.ArgumentParser()
parser.add_argument('--gpu', default="-1", type=str, nargs='?', help='gpu id')
parser.add_argument('--dataset', default="german", type=str, nargs='?', help='dataset')
parser.add_argument('--sensitive', default="age", type=str, nargs='?', help='sensitive feature')
parser.add_argument('--validation_size', default="0.05", type=str, nargs='?', help='validation size')
parser.add_argument('--test_size', default="0.05", type=str, nargs='?', help='test size')
parser.add_argument('--n_run', default="5", type=str, nargs='?', help='no of running times')
args = parser.parse_args()
print("gpu_id: {}, dataset: {}, sensitive: {}, validation_size: {}, test_size: {}, n_run: {}".
format(args.gpu, args.dataset, args.sensitive, args.validation_size, args.test_size, args.n_run))
os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
K.set_session(sess)
dataset = args.dataset
sensitive = args.sensitive
valid_size = float(args.validation_size)
test_size = float(args.test_size)
n_run = int(args.n_run)
save_folder_model = "initial_model"
### hyper-parameters of NN
hidden_dim = 32
batch_size = 32
epochs = 20
start_date_time = datetime.datetime.now()
start_time = timeit.default_timer()
# read data
X_data, y_data, n_data, n_feature, sen_var_indices = read_data.from_file(dataset, sensitive)
# get sensitive feature index
sen_idx = sen_var_indices[0]
# normalize X
X_data = MinMaxScaler().fit_transform(X_data)
# accuracy, group_fairness, individual_fairness
acc_valid_initial, fair_valid_initial, individual_valid_initial = np.zeros(n_run), np.zeros(n_run), np.zeros(n_run)
acc_test_initial, fair_test_initial, individual_test_initial = np.zeros(n_run), np.zeros(n_run), np.zeros(n_run)
for run in range(n_run):
print("run={}".format(run))
# split data to training, validation, and testing sets
X_train_test, X_valid, y_train_test, y_valid = train_test_split(X_data, y_data, test_size=valid_size,
random_state=run, shuffle=True)
X_train, X_test, y_train, y_test = train_test_split(X_train_test, y_train_test, test_size=test_size,
random_state=run, shuffle=True)
print("n_train: {}, n_valid: {}, n_test: {}".format(len(y_train), len(y_valid), len(y_test)))
# train NN once to obtain initial function
time1 = timeit.default_timer()
trained_model = Sequential()
trained_model.add(Dense(hidden_dim, input_dim=n_feature, activation='relu'))
trained_model.add(Dense(1, activation='sigmoid'))
trained_model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
trained_model.fit(X_train, y_train, shuffle=True, epochs=epochs, batch_size=batch_size, verbose=0)
# compute accuracy and fairness on validation set
y_pred_validation = trained_model.predict(X_valid)
y_pred_validation_round = np.around(y_pred_validation)
accuracy_overall_valid, demographic_parity_valid, prob_favored_pred_positive_valid, prob_unfavored_pred_positive_valid \
= group_fairness.compute_accuracy_fairness(X_valid, sen_idx, y_valid, y_pred_validation_round)
print("initial func on validation: accuracy={}, fairness={}, p_favored_positive={}, p_unfavored_positive={}".
format(round(accuracy_overall_valid, 2), round(demographic_parity_valid, 2),
round(prob_favored_pred_positive_valid, 4), round(prob_unfavored_pred_positive_valid, 4)))
theil_index_valid = individual_fairness.generalized_entropy_index(y_valid, y_pred_validation_round)
print("initial func on validation: theil_index={}".format(round(theil_index_valid, 2)))
acc_valid_initial[run] = accuracy_overall_valid
fair_valid_initial[run] = demographic_parity_valid
individual_valid_initial[run] = theil_index_valid
# save validation set and prediction to file
with open("./{}/X_valid_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, X_valid)
with open("./{}/y_valid_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, y_valid)
with open("./{}/y_pred_validation_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, y_pred_validation)
# compute accuracy and fairness on testing set
y_pred_testing = trained_model.predict(X_test)
y_pred_testing_round = np.around(y_pred_testing)
accuracy_overall_test, demographic_parity_test, prob_favored_pred_positive_test, prob_unfavored_pred_positive_test \
= group_fairness.compute_accuracy_fairness(X_test, sen_idx, y_test, y_pred_testing_round)
print("initial func on testing: accuracy={}, fairness={}, p_favored_positive={}, p_unfavored_positive={}".
format(round(accuracy_overall_test, 2), round(demographic_parity_test, 2),
round(prob_favored_pred_positive_test, 4), round(prob_unfavored_pred_positive_test, 4)))
theil_index_test = individual_fairness.generalized_entropy_index(y_test, y_pred_testing_round)
print("initial func on testing: theil_index={}".format(round(theil_index_test, 2)))
acc_test_initial[run] = accuracy_overall_test
fair_test_initial[run] = demographic_parity_test
individual_test_initial[run] = theil_index_test
# save testing set and prediction to file
with open("./{}/X_test_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, X_test)
with open("./{}/y_test_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, y_test)
with open("./{}/y_pred_testing_{}_{}_vs{}_ts{}_run{}.file".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run), "wb") as f:
np.save(f, y_pred_testing)
# save NN
trained_model.save("./{}/model_{}_{}_vs{}_ts{}_run{}.h5".
format(save_folder_model, dataset, sensitive, valid_size, test_size, run))
time2 = timeit.default_timer()
print("runtime of NN: {}(s)".format(round(time2 - time1, 2)))
del trained_model
K.clear_session()
# end run
end_date_time = datetime.datetime.now()
end_time = timeit.default_timer()
print("start date time: {} and end date time: {}".format(start_date_time, end_date_time))
print("runtime: {}(s)".format(round(end_time-start_time, 2)))
# save result to file
acc_valid, acc_valid_std = round(np.mean(acc_valid_initial), 2), round(np.std(acc_valid_initial), 2)
fair_valid, fair_valid_std = round(np.mean(fair_valid_initial), 2), round(np.std(fair_valid_initial), 2)
individual_valid, individual_valid_std = round(np.mean(individual_valid_initial), 2), round(np.std(individual_valid_initial), 2)
acc_test, acc_test_std = round(np.mean(acc_test_initial), 2), round(np.std(acc_test_initial), 2)
fair_test, fair_test_std = round(np.mean(fair_test_initial), 2), round(np.std(fair_test_initial), 2)
individual_test, individual_test_std = round(np.mean(individual_test_initial), 2), round(np.std(individual_test_initial), 2)
with open('./{}/_model_{}_{}_vs{}_ts{}.txt'.format(save_folder_model, dataset, sensitive, valid_size, test_size), 'w') as f:
f.write("method: initial, dataset: {}, sensitive: {}\n".format(dataset, sensitive))
f.write("n_train: {}, n_valid: {}, n_test: {}\n".format(len(y_train), len(y_valid), len(y_test)))
f.write("acc_valid: {} ({}), fair_valid: {} ({}), individual_valid: {} ({})\n".
format(acc_valid, acc_valid_std, fair_valid, fair_valid_std, individual_valid, individual_valid_std))
f.write("acc_test: {} ({}), fair_test: {} ({}), individual_test: {} ({})\n".
format(acc_test, acc_test_std, fair_test, fair_test_std, individual_test, individual_test_std))
f.write("start date time: {} and end date time: {}\n".format(start_date_time, end_date_time))
f.write("runtime: {}(s)\n".format(round(end_time-start_time, 2)))