-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathconfig_3dmatch.py
216 lines (171 loc) · 8.72 KB
/
config_3dmatch.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
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
import argparse
arg_lists = []
parser = argparse.ArgumentParser()
def add_argument_group(name):
arg = parser.add_argument_group(name)
arg_lists.append(arg)
return arg
def str2bool(v):
return v.lower() in ('true', '1')
logging_arg = add_argument_group('Logging')
trainer_arg = add_argument_group('Trainer')
trainer_arg.add_argument('--trainer', type=str, default='HardestContrastiveLossTrainer')
trainer_arg.add_argument('--save_freq_epoch', type=int, default=1)
trainer_arg.add_argument('--batch_size', type=int, default=2)
trainer_arg.add_argument('--val_batch_size', type=int, default=1)
# Hard negative mining
trainer_arg.add_argument('--use_hard_negative', type=str2bool, default=True)
trainer_arg.add_argument('--hard_negative_sample_ratio', type=int, default=0.05)
trainer_arg.add_argument('--hard_negative_max_num', type=int, default=3000)
trainer_arg.add_argument('--num_pos_per_batch', type=int, default=1024)
trainer_arg.add_argument('--num_hn_samples_per_batch', type=int, default=256)
# Metric learning loss
trainer_arg.add_argument('--neg_thresh', type=float, default=1.4)
trainer_arg.add_argument('--pos_thresh', type=float, default=0.1)
trainer_arg.add_argument('--neg_weight', type=float, default=1)
# Data augmentation
trainer_arg.add_argument('--use_random_scale', type=str2bool, default=False)
trainer_arg.add_argument('--min_scale', type=float, default=0.8)
trainer_arg.add_argument('--max_scale', type=float, default=1.2)
trainer_arg.add_argument('--use_random_rotation', type=str2bool, default=True)
trainer_arg.add_argument('--rotation_range', type=float, default=360)
# Data loader configs
trainer_arg.add_argument('--train_phase', type=str, default="train")
trainer_arg.add_argument('--val_phase', type=str, default="val")
trainer_arg.add_argument('--test_phase', type=str, default="test")
trainer_arg.add_argument('--stat_freq', type=int, default=40)
trainer_arg.add_argument('--test_valid', type=str2bool, default=True)
trainer_arg.add_argument('--val_max_iter', type=int, default=400)
trainer_arg.add_argument('--val_epoch_freq', type=int, default=1)
trainer_arg.add_argument(
'--positive_pair_search_voxel_size_multiplier', type=float, default=1.5)
trainer_arg.add_argument('--hit_ratio_thresh', type=float, default=0.1)
# Triplets
trainer_arg.add_argument('--triplet_num_pos', type=int, default=256)
trainer_arg.add_argument('--triplet_num_hn', type=int, default=512)
trainer_arg.add_argument('--triplet_num_rand', type=int, default=1024)
# dNetwork specific configurations
net_arg = add_argument_group('Network')
net_arg.add_argument('--model', type=str, default='ResUNetBN2C')
net_arg.add_argument('--model_n_out', type=int, default=32, help='Feature dimension')
net_arg.add_argument('--conv1_kernel_size', type=int, default=5)
net_arg.add_argument('--normalize_feature', type=str2bool, default=True)
net_arg.add_argument('--dist_type', type=str, default='L2')
net_arg.add_argument('--best_val_metric', type=str, default='feat_match_ratio',help='[feat_match_ratio,rre,rte]')
# Optimizer arguments
opt_arg = add_argument_group('Optimizer')
opt_arg.add_argument('--optimizer', type=str, default='SGD')
opt_arg.add_argument('--max_epoch', type=int, default=200)
opt_arg.add_argument('--lr', type=float, default=1e-1)
opt_arg.add_argument('--momentum', type=float, default=0.8)
opt_arg.add_argument('--sgd_momentum', type=float, default=0.9)
opt_arg.add_argument('--sgd_dampening', type=float, default=0.1)
opt_arg.add_argument('--adam_beta1', type=float, default=0.9)
opt_arg.add_argument('--adam_beta2', type=float, default=0.999)
opt_arg.add_argument('--weight_decay', type=float, default=1e-4)
opt_arg.add_argument('--iter_size', type=int, default=1, help='accumulate gradient')
opt_arg.add_argument('--bn_momentum', type=float, default=0.05)
opt_arg.add_argument('--exp_gamma', type=float, default=0.99)
opt_arg.add_argument('--scheduler', type=str, default='ExpLR')
icp_path = "/DISK/qwt/datasets/kitti/data_odometry_velodyne/dataset/icp"
opt_arg.add_argument('--icp_cache_path', type=str, default=icp_path)
misc_arg = add_argument_group('Misc')
misc_arg.add_argument('--use_gpu', type=str2bool, default=True)
misc_arg.add_argument('--weights', type=str, default=None)
misc_arg.add_argument('--weights_dir', type=str, default=None)
misc_arg.add_argument('--resume', type=str, default=None)
misc_arg.add_argument('--teacher_weight', type=str, default='/home/ymz/桌面/Distillation/3DMatch.pth')
misc_arg.add_argument('--fast_validation', type=str2bool, default=False)
misc_arg.add_argument('--nn_max_n',
type=int,
default=500,
help='The maximum number of features to find nearest neighbors in batch')
# Dataset specific configurations
data_arg = add_argument_group('Data')
# ----------------------------------------------------------------------- #
# 3DMatch ---- |output path|
output_3DMatch = "../outputs"
logging_arg.add_argument('--out_dir', type=str, default=output_3DMatch)
# 3DMatch ---- |resume dir|
misc_arg.add_argument('--resume_dir', type=str, default=None)
# 3DMtach ---- |num thread|
misc_arg.add_argument('--train_num_thread', type=int, default=8)
misc_arg.add_argument('--val_num_thread', type=int, default=8)
misc_arg.add_argument('--test_num_thread', type=int, default=8)
# 3DMatch ---- |dataset|
dataset_3DMatch = 'ThreeDMatchPairDataset'
data_arg.add_argument('--dataset', type=str, default=dataset_3DMatch)
# 3DMatch ---- |voxel size|
voxel_size_3DMatch = 0.025
# voxel_size_3DMatch = 0.05
data_arg.add_argument('--voxel_size', type=float, default=voxel_size_3DMatch)
# ----------------------------------------------------------------------- #
# Dataset path
# data_path = "/DISK/qwt/datasets/Ours_train_0_01/train"
data_path = "/media/ymz/2b933929-0294-4162-9385-4fe3eec72189/distillation/3DImageMatch/3DImageMatch/train"
data_arg.add_argument('--threed_match_dir', type=str, default=data_path)
# overlap_path = "/DISK/qwt/datasets/Ours_train_0_01/overlap"
overlap_path = '/media/ymz/2b933929-0294-4162-9385-4fe3eec72189/distillation/3DImageMatch/3DImageMatch/overlap'
data_arg.add_argument('--overlap_path', type=str, default=overlap_path)
# image setting
data_arg.add_argument('--image_W', type=str, default=160)
data_arg.add_argument('--image_H', type=str, default=120)
kitti_path = "/DISK/qwt/datasets/kitti/data_odometry_velodyne"
data_arg.add_argument('--kitti_root', type=str, default=kitti_path)
data_arg.add_argument(
'--kitti_max_time_diff',
type=int,
default=3,
help='max time difference between pairs (non inclusive)')
data_arg.add_argument('--kitti_date', type=str, default='2020_09_30')
#kpconv
architectures = dict()
kpfcn_backbone = [
'simple',
'resnetb',
'resnetb_strided',
'resnetb',
'resnetb',
'resnetb_strided',
'resnetb',
'resnetb',
'resnetb_strided',
'resnetb',
'last_unary',
# 'nearest_upsample',
# 'unary',
# 'nearest_upsample',
# 'unary',
# 'nearest_upsample',
# 'last_unary'
]
architectures['KPFCN'] = kpfcn_backbone
KP_arg = add_argument_group('KPConv')
KP_arg.add_argument('--architecture', type=list, default=kpfcn_backbone)
KP_arg.add_argument('--num_layers', type=int, default=4)
KP_arg.add_argument('--in_points_dim', type=int, default=3)
KP_arg.add_argument('--first_feats_dim', type=int, default=128)
KP_arg.add_argument('--final_feats_dim', type=int, default=32)
KP_arg.add_argument('--first_subsampling_dl', type=float, default=0.025)
KP_arg.add_argument('--in_feats_dim', type=int, default=1)
KP_arg.add_argument('--conv_radius', type=float, default=2.5)
KP_arg.add_argument('--deform_radius', type=float, default=5.0)
KP_arg.add_argument('--num_kernel_points', type=int, default=15)
KP_arg.add_argument('--KP_extent', type=float, default=2.0)
KP_arg.add_argument('--batch_norm_momentum', type=float, default=0.02)
KP_arg.add_argument('--use_batch_norm', type=bool, default=True)
KP_arg.add_argument('--add_cross_score', type=bool, default=True)
KP_arg.add_argument('--condition_feature', type=bool, default=True)
KP_arg.add_argument('--deformable', type=bool, default=False)
KP_arg.add_argument('--modulated', type=bool, default=False)
KP_arg.add_argument('--KP_influence', type=str, default='linear')
KP_arg.add_argument('--aggregation_mode', type=str, default='sum')
KP_arg.add_argument('--fixed_kernel_points', type=str, default='center')
Distillation = add_argument_group('Distillation')
Distillation.add_argument('--w_kd', type=float, default=0.0, help='weight for distillation loss')
Distillation.add_argument('--initial_fcgf', type=bool, default=True, help='initalize the weight of fcgf backbone')
Distillation.add_argument('--temperature', type=float, default=1.0)
Distillation.add_argument('--use_softmax', type=bool, default=True, help='use softmax in kd')
def get_config():
args = parser.parse_args()
return args