-
Notifications
You must be signed in to change notification settings - Fork 3
/
opts.py
145 lines (140 loc) · 5.58 KB
/
opts.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 argparse
parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
parser.add_argument(
'data',
metavar='DIR',
default='/data-sets/imagenet-eureka/imagenet256/',
help='path to dataset')
parser.add_argument('-j', '--workers', default=16, type=int, metavar='N',
help='number of data loading workers (default: 4)')
parser.add_argument('--epochs', default=90, type=int, metavar='N',
help='number of total epochs to run')
parser.add_argument('--start-epoch', default=0, type=int, metavar='N',
help='manual epoch number (useful on restarts)')
parser.add_argument('-b', '--batch-size', default=256, type=int,
metavar='N',
help='mini-batch size (default: 256), this is the total '
'batch size of all GPUs on the current node when '
'using Data Parallel or Distributed Data Parallel')
parser.add_argument('--lr', '--learning-rate', default=0.001, type=float,
metavar='LR', help='initial learning rate', dest='lr')
parser.add_argument('--momentum', default=0.9, type=float, metavar='M',
help='momentum')
parser.add_argument('--wd', '--weight-decay', default=1e-2, type=float,
metavar='W', help='weight decay (default: 1e-2)',
dest='weight_decay')
parser.add_argument('-p', '--print-freq', default=10, type=int,
metavar='N', help='print frequency (default: 10)')
parser.add_argument('--resume', default='', type=str, metavar='PATH',
help='path to latest checkpoint (default: none)')
parser.add_argument('-e', '--evaluate', dest='evaluate', action='store_true',
help='evaluate model on validation set')
parser.add_argument('--world-size', default=-1, type=int,
help='number of nodes for distributed training')
parser.add_argument('--rank', default=-1, type=int,
help='node rank for distributed training')
parser.add_argument('--dist-url', default='tcp://127.0.0.1:23456', type=str,
help='url used to set up distributed training')
parser.add_argument('--dist-backend', default='nccl', type=str,
help='distributed backend')
parser.add_argument('--seed', default=None, type=int,
help='seed for initializing training. ')
parser.add_argument('--gpu', default=None, type=int,
help='GPU id to use.')
parser.add_argument(
'--multiprocessing-distributed',
action='store_true',
help='Use multi-processing distributed training to launch '
'N processes per node, which has N GPUs. This is the '
'fastest way to use PyTorch for either single node or '
'multi node data parallel training')
parser.add_argument(
'--optimizer',
type=str,
choices=[
'sgd',
'adam',
'adamw'],
default='adamw')
parser.add_argument('--use-mixup', action='store_true')
parser.add_argument('--alpha', type=float, default=0.2)
parser.add_argument('--output-dir', type=str, default='output-exp/')
parser.add_argument(
'--scheduler',
type=str,
choices=[
'multistep',
'cosine'],
default='multistep')
parser.add_argument('--warmup', type=int, default=0)
parser.add_argument('--stem-type', type=str, default='basic')
parser.add_argument('--resume-epoch', action='store_true',
help='if enabled, will resume the epoch')
# Experts
parser.add_argument('--num-experts', type=int, default=4)
parser.add_argument('--use-only-first', action='store_true')
parser.add_argument(
'--expansion-stage',
action='store_true',
help='If enabled it will replicate the weights from expert 0 to the rest.')
# Binarization
parser.add_argument(
'--binary-activations',
action='store_true',
help='Binarize the activations')
parser.add_argument(
'--binary-weights',
action='store_true',
help='Binarize the weights')
# Network structure
parser.add_argument('--structure', nargs='+', type=int, default=[2, 2, 2, 2])
parser.add_argument('--num-groups', nargs='+', type=int, default=[1, 1, 1, 1])
parser.add_argument('--expansion', nargs='+', type=float, default=[1, 1, 1, 1])
parser.add_argument('--downsample-ratio', type=int, default=4)
parser.add_argument(
'--add-g-layer',
action='store_true',
help='if true, for g>1 a 1x1 layer will be added')
parser.add_argument('--use-se', action='store_true', help='Add a SE layer')
# distillation
parser.add_argument('--teacher', type=str, default='', help='teacher weights')
parser.add_argument(
'--teacher-config',
type=str,
default='',
help='path to Json containing the args to build the teacher model.')
parser.add_argument(
'--att-transfer',
default=False,
action='store_true',
help='Do attention transfer from real-valued network?')
parser.add_argument(
'--att-transfer-weighting',
default=1e+3,
type=float,
help='weighting of the att. transfer terms within the loss')
parser.add_argument(
'--att-transfer-indicator',
default=[
0,
1,
1,
1],
type=int,
nargs="+",
help="Which stages to use for attention transfer")
parser.add_argument(
'--lab-match',
default=False,
action='store_true',
help='Match soft labels of the teacher network?')
parser.add_argument(
'--lab-match-w',
default=1e+3,
type=float,
help='weighting of the soft label matching loss term')
parser.add_argument(
'--lab-match-T',
default=1,
type=float,
help='Temperature to apply when computing the labels of the teacher')