-
Notifications
You must be signed in to change notification settings - Fork 71
/
trainval_config_i.py
170 lines (164 loc) · 5.5 KB
/
trainval_config_i.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
dataset_type = 'KittiDataset'
data_root = '../../../../data/DAIR-V2X/cooperative-vehicle-infrastructure/infrastructure-side/'
class_names = ['Car']
input_modality = dict(use_lidar=False, use_camera=True)
point_cloud_range = [0, -39.68, -3, 92.16, 39.68, 1]
voxel_size = [0.32, 0.32, 0.33]
length = int((point_cloud_range[3] - point_cloud_range[0]) / voxel_size[0])
width = int((point_cloud_range[4] - point_cloud_range[1]) / voxel_size[1])
height = int((point_cloud_range[5] - point_cloud_range[2]) / voxel_size[2])
output_shape = [width, length, height]
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
img_scale = (960, 540)
img_resize_scale = [(912, 513), (1008, 567)]
work_dir = './work_dirs/vic3d_latefusion_inf_imvoxelnet'
model = dict(
type='ImVoxelNet',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50'),
style='pytorch'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=64,
num_outs=4),
neck_3d=dict(type='OutdoorImVoxelNeck', in_channels=64, out_channels=256),
bbox_head=dict(
type='Anchor3DHead',
num_classes=1,
in_channels=256,
feat_channels=256,
use_direction_classifier=True,
anchor_generator=dict(
type='AlignedAnchor3DRangeGenerator',
ranges=[[0, -39.68, -1.78, 92.16, 39.68, -1.78]],
sizes=[[3.9, 1.6, 1.56]],
rotations=[0, 1.57],
reshape_out=True),
diff_rad_by_sin=True,
bbox_coder=dict(type='DeltaXYZWLHRBBoxCoder'),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, loss_weight=2.0),
loss_dir=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.2)),
n_voxels=output_shape,
anchor_generator=dict(
type='AlignedAnchor3DRangeGenerator',
ranges=[[0, -39.68, -3.08, 92.16, 39.68, 0.76]],
rotations=[.0]),
train_cfg=dict(
assigner=dict(
type='MaxIoUAssigner',
iou_calculator=dict(type='BboxOverlapsNearest3D'),
pos_iou_thr=0.6,
neg_iou_thr=0.45,
min_pos_iou=0.45,
ignore_iof_thr=-1),
allowed_border=0,
pos_weight=-1,
debug=False),
test_cfg=dict(
use_rotate_nms=True,
nms_across_levels=False,
nms_thr=0.01,
score_thr=0.1,
min_bbox_size=0,
nms_pre=100,
max_num=50))
train_pipeline = [
dict(type='LoadAnnotations3D'),
dict(type='LoadImageFromFile'),
dict(type='RandomFlip3D', flip_ratio_bev_horizontal=0.5),
dict(
type='Resize',
img_scale=img_resize_scale,
keep_ratio=True,
multiscale_mode='range'),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='ObjectRangeFilter', point_cloud_range=point_cloud_range),
dict(type='DefaultFormatBundle3D', class_names=class_names),
dict(type='Collect3D', keys=['img', 'gt_bboxes_3d', 'gt_labels_3d'])
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='Resize', img_scale=img_scale, keep_ratio=True),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(
type='DefaultFormatBundle3D',
class_names=class_names,
with_label=False),
dict(type='Collect3D', keys=['img'])
]
data = dict(
samples_per_gpu=1,
workers_per_gpu=1,
train=dict(
type='RepeatDataset',
times=1,
dataset=dict(
type=dataset_type,
data_root=data_root,
ann_file=data_root + 'kitti_infos_train.pkl',
split='training',
pts_prefix='velodyne_reduced',
pipeline=train_pipeline,
modality=input_modality,
classes=class_names,
test_mode=False)),
val=dict(
type=dataset_type,
data_root=data_root,
ann_file=data_root + 'kitti_infos_val.pkl',
split='training',
pts_prefix='velodyne_reduced',
pipeline=test_pipeline,
modality=input_modality,
classes=class_names,
test_mode=True),
test=dict(
type=dataset_type,
data_root=data_root,
ann_file=data_root + 'kitti_infos_val.pkl',
split='training',
pts_prefix='velodyne_reduced',
pipeline=test_pipeline,
modality=input_modality,
classes=class_names,
box_type_3d="Lidar",
test_mode=True))
optimizer = dict(
type='AdamW',
lr=0.0001,
weight_decay=0.0001,
paramwise_cfg=dict(
custom_keys={'backbone': dict(lr_mult=0.1, decay_mult=1.0)}))
optimizer_config = dict(grad_clip=dict(max_norm=35., norm_type=2))
lr_config = dict(policy='step', step=[8, 11])
total_epochs = 12
checkpoint_config = dict(interval=1, max_keep_ckpts=1)
log_config = dict(
interval=50,
hooks=[dict(type='TextLoggerHook'),
dict(type='TensorboardLoggerHook')])
evaluation = dict(interval=1)
dist_params = dict(backend='nccl')
find_unused_parameters = True # only 1 of 4 FPN outputs is used
log_level = 'INFO'
load_from = None
resume_from = None
workflow = [('train', 1)]