-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
178 lines (145 loc) · 5.72 KB
/
train.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
import os
import argparse
import json
import datetime
import numpy as np
import random
import time
from tqdm import tqdm
import torch
from torch.utils.data import DataLoader
from torch.optim import SGD, Optimizer
from torch.optim.lr_scheduler import StepLR
from torch.nn import CrossEntropyLoss
import torch.nn as nn
from sklearn.metrics import (accuracy_score, balanced_accuracy_score,
confusion_matrix, f1_score)
from datasets import AdultDataset,CompasDataset,CrimeDataset,GermanDataset,HealthDataset,LawschoolDataset
import models
from utils import fairness_metric
from utils import Averagemeter
parser = argparse.ArgumentParser(description='Fairness Training')
parser.add_argument('-dataset', type=str, choices=['adult','compas','crime','german','health','lawschool'], default='compas')
parser.add_argument('-data_dir', type=str, default='./data/')
parser.add_argument('-protected_att', type=str)
parser.add_argument('-load', type=str)
parser.add_argument('-label', type=str)
parser.add_argument('-transfer', type=bool, default=False)
parser.add_argument('-quantiles', type=bool, default=True)
parser.add_argument('-num_epochs', type=int, default=100)
parser.add_argument('-batch-size', type=int, default=256)
parser.add_argument('-balanced', type=bool, default=False)
# crime: 0.05
# German: 0.05
# others: 0.001
parser.add_argument('-learning_rate', type=float, default=0.001)
parser.add_argument('-weight_decay', type=float, default=0.01)
parser.add_argument('-patience', type=float, default=5)
parser.add_argument('-num_models', type=int, default=1)
args = parser.parse_args()
# device
USE_CUDA = torch.cuda.is_available()
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# init seed
def set_seed(SEED):
np.random.seed(SEED)
torch.manual_seed(SEED)
random.seed(SEED)
if USE_CUDA:
torch.cuda.manual_seed_all(SEED)
# dataset
if args.dataset=='adult':
train_dataset = AdultDataset('train', args)
val_dataset = AdultDataset('validation', args)
test_dateset = AdultDataset('test', args)
elif args.dataset == 'compas':
train_dataset = CompasDataset('train', args)
val_dataset = CompasDataset('validation', args)
test_dateset = CompasDataset('test', args)
elif args.dataset == 'crime':
train_dataset = CrimeDataset('train', args)
val_dataset = CrimeDataset('validation', args)
test_dateset = CrimeDataset('test', args)
elif args.dataset == 'german':
train_dataset = GermanDataset('train', args)
val_dataset = GermanDataset('validation', args)
test_dateset = GermanDataset('test', args)
elif args.dataset == 'health':
train_dataset = HealthDataset('train', args)
val_dataset = HealthDataset('validation', args)
test_dateset = HealthDataset('test', args)
elif args.dataset == 'lawschool':
train_dataset = LawschoolDataset('train', args)
val_dataset = LawschoolDataset('validation', args)
test_dateset = LawschoolDataset('test', args)
train_loader = torch.utils.data.DataLoader(
train_dataset, batch_size=args.batch_size, shuffle=True
)
val_loader = torch.utils.data.DataLoader(
val_dataset, batch_size=args.batch_size, shuffle=False
)
test_loader = torch.utils.data.DataLoader(
test_dateset, batch_size=args.batch_size, shuffle=False
)
def run(model, optimizer, loader, split, epoch):
predictions = []
targets = []
protects = []
logits_all = []
avg_loss = Averagemeter.AverageMeter('avg_loss')
for data, target, protected in loader:
data, target, protected = data.to(device), target.to(device), protected.to(device)
logits = model(data).squeeze()
bce_loss = binary_cross_entropy(logits, target)
if split == 'train':
optimizer.zero_grad()
bce_loss.backward()
optimizer.step()
logits_all.append(logits.cpu())
prediction = (0.5 <= nn.Sigmoid().to(device)(logits)).float().squeeze()
predictions.append(prediction.detach().cpu())
targets.append(target.detach().cpu())
protects.append(protected.cpu())
avg_loss.update(bce_loss.mean().item(), len(data))
predictions = torch.cat(predictions)
targets = torch.cat(targets)
protects = torch.cat(protects)
logits_all = torch.cat(logits_all)
# if split == 'train' and epoch == args.num_epochs-1:
# torch.save(targets,'./logs/labels')
# torch.save(logits_all,'./logs/logits')
# torch.save(protects,'./logs/protects')
if split == 'test':
accuracy = accuracy_score(targets, predictions)
print(f'Acc: {accuracy}')
parity = fairness_metric.statistical_parity(predictions,protects)
print(f'Parity: {parity}')
print(f'Test loss: {avg_loss.avg}')
return avg_loss.avg
# model preparation
# 200,20
print('length of features')
print(train_dataset.features.shape[1])
model = models.MLP([train_dataset.features.shape[1], 20, 20, 1])
# training preparation
binary_cross_entropy = nn.BCEWithLogitsLoss(
pos_weight=train_dataset.pos_weight('train') if args.balanced else None
)
optimizer = torch.optim.Adam(
list(model.parameters()),
lr=args.learning_rate, weight_decay=args.weight_decay
)
scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(
optimizer, 'min', patience=args.patience, factor=0.5
)
# train the model
for epoch in tqdm(range(args.num_epochs)):
model.train()
loss_train = run(model, optimizer, train_loader, 'train', epoch)
# print('epoch {}: lr {} loss {}'.format(epoch,optimizer.state_dict()['param_groups'][0]['lr'],loss_train))
model.eval()
loss_val = run(model, optimizer, val_loader, 'valid', epoch)
scheduler.step(loss_val)
torch.save(model,f'logs/{args.dataset}_model')
# test the model
loss_test = run(model, optimizer, test_loader, 'test', 0)