-
Notifications
You must be signed in to change notification settings - Fork 17
/
Copy pathcollect_results.py
136 lines (123 loc) · 6.6 KB
/
collect_results.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
import argparse
import collections
import json
import os
import numpy as np
import dataset
import mlconfig
import models
import torch
import util
from evaluator import Evaluator
from tabulate import tabulate
parser = argparse.ArgumentParser(description='ClasswiseNoise')
args = parser.parse_args()
if torch.cuda.is_available():
torch.backends.cudnn.enabled = True
torch.backends.cudnn.benchmark = True
device = torch.device('cuda')
device_list = [torch.cuda.get_device_name(i) for i in range(0, torch.cuda.device_count())]
print("GPU List: %s" % (device_list))
else:
device = torch.device('cpu')
print("PyTorch Version: %s" % (torch.__version__))
def load_results(targt_exp, model_name):
# print(targt_exp)
config_file = os.path.join(targt_exp, model_name+'.yaml')
checkpoint_path_file = os.path.join(targt_exp, 'checkpoints', model_name)
if not os.path.isfile(config_file) or not os.path.isfile(checkpoint_path_file+'.pth'):
# print('No such files: \n%s\n%s' % (config_file, checkpoint_path_file))
return None
config = mlconfig.load(config_file)
config.set_immutable()
model = config.model().to(device)
checkpoints = util.load_model(filename=checkpoint_path_file, model=model, optimizer=None, scheduler=None)
if config.epochs != checkpoints['epoch']:
return None
if 'cm_history' in checkpoints['ENV']:
new_hist = []
for item in checkpoints['ENV']['cm_history']:
if isinstance(item, np.ndarray):
new_hist.append(item.tolist())
else:
new_hist.append(item)
checkpoints['ENV']['cm_history'] = new_hist
return checkpoints['ENV']
if __name__ == '__main__':
exp_names = [
'experiments/cifar10/random_samplewise/CIFAR10-eps=8',
'experiments/cifar10/min-max_samplewise/CIFAR10-eps=8-se=0.9-base_version=resnet18',
'experiments/cifar10/min-min_samplewise/CIFAR10-eps=8-se=0.1-base_version=resnet18',
'experiments/cifar10/min-min_samplewise/CIFAR10-eps=8-se=0.01-base_version=resnet18',
'experiments/cifar100/min-min_samplewise/CIFAR100-eps=8-se=0.3-base_version=resnet18',
'experiments/cifar100/min-min_samplewise/CIFAR100-eps=8-se=0.01-base_version=resnet18',
'experiments/svhn/min-min_samplewise/SVHN-eps=8-se=0.1-base_version=resnet18',
'experiments/svhn/min-min_samplewise/SVHN-eps=8-se=0.01-base_version=resnet18',
'experiments/imagenet-mini/min-min_samplewise/ImageNetMini-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar10/random_classwise/CIFAR10-eps=8/',
'experiments/cifar10/min-max_classwise/CIFAR10-eps=8-se=0.8-base_version=resnet18',
'experiments/cifar10/min-min_classwise/CIFAR10-eps=8-se=0.1-base_version=resnet18',
'experiments/cifar10/min-min_classwise/CIFAR10-eps=8-se=0.01-base_version=resnet18',
'experiments/cifar100/min-min_classwise/CIFAR100-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar100/min-min_classwise/CIFAR100-eps=8-se=0.01-base_version=resnet18',
'experiments/svhn/min-min_classwise/SVHN-eps=8-se=0.1-base_version=resnet18',
'experiments/svhn/min-min_classwise/SVHN-eps=8-se=0.01-base_version=resnet18',
'experiments/imagenet-mini/min-min_classwise/ImageNetMini-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=24-se=0.01-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=24-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=24-se=0.01-base_version=resnet18',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=16-se=0.01-base_version=resnet18',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=24-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=24-se=0.01-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=8-se=0.1-base_version=resnet18-2noise',
'experiments/cifar10-extension/min-min_classwise/TinyImageNet-eps=16-se=0.1-base_version=resnet18',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random8',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random16',
'experiments/cifar10-extension/min-min_classwise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random24',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random8',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random16',
'experiments/cifar10-extension/min-min_samplewise/CIFAR10-eps=8-se=0.1-base_version=resnet18-random24',
]
model_list = [
'resnet18',
'resnet50',
'dense121',
'resnet18_augmentation',
'resnet18_madrys',
'resnet18_classpoison',
'resnet18_classpoison_targeted',
'resnet18_add-uniform-noise',
'resnet18_add-uniform-noise-aug',
'resnet18_cutout',
'resnet18_cutmix',
'resnet18_mixup',
]
poison_rate_list = [0.0, 0.1, 0.2, 0.4, 0.6, 0.8, 1.0]
exp_results = {}
for exp_name in exp_names:
print(exp_name)
table_data_header = ['Model'] + poison_rate_list
table_data = [model_list]
exp_results[exp_name] = {}
for poison_rate in poison_rate_list:
target_dir = os.path.join(exp_name, 'poison_train_%.1f' % poison_rate)
temp_list = []
exp_results[exp_name][poison_rate] = {}
for model_name in model_list:
rs_env = load_results(os.path.join(target_dir, model_name), model_name)
exp_results[exp_name][poison_rate][model_name] = rs_env
if rs_env is not None:
temp_list.append('%.2f' % rs_env['curren_acc'])
else:
temp_list.append('..')
table_data.append(temp_list)
# Transpose array
table_data = list(map(list, zip(*table_data)))
print('=' * 40 + 'Results' + '=' * 40)
print(tabulate(table_data, headers=table_data_header, floatfmt=".2f", stralign="left", numalign="left"))
print('=' * (80 + len('Results')) + '\n')
# Save results to
with open('exp_results.json', 'w') as outfile:
json.dump(exp_results, outfile)