-
Notifications
You must be signed in to change notification settings - Fork 29
/
dino_4scale_flash_internimage_t_1x_coco.py
179 lines (178 loc) · 6.46 KB
/
dino_4scale_flash_internimage_t_1x_coco.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
# --------------------------------------------------------
# DCNv4
# Copyright (c) 2023 OpenGVLab
# Licensed under The MIT License [see LICENSE for details]
# --------------------------------------------------------
_base_ = [
'../_base_/datasets/coco_detection.py',
'../_base_/default_runtime.py',
'../_base_/schedules/schedule_1x.py',
]
pretrained = 'https://huggingface.co/OpenGVLab/DCNv4/resolve/main/flash_intern_image_t_1k_224.pth'
model = dict(
type='DINO',
backbone=dict(
type='FlashInternImage',
core_op='DCNv4',
channels=64,
depths=[4, 4, 18, 4],
groups=[4, 8, 16, 32],
mlp_ratio=4.,
drop_path_rate=0.2,
norm_layer='LN',
layer_scale=1.0,
offset_scale=1.0,
post_norm=False,
with_cp=True,
out_indices=(1, 2, 3),
init_cfg=dict(type='Pretrained', checkpoint=pretrained)),
neck=dict(
type='ChannelMapper',
in_channels=[128, 256, 512],
kernel_size=1,
out_channels=256,
act_cfg=None,
norm_cfg=dict(type='GN', num_groups=32),
num_outs=4),
bbox_head=dict(
type='DINOHead',
num_query=900,
num_classes=80,
in_channels=2048,
sync_cls_avg_factor=True,
as_two_stage=True,
with_box_refine=True,
dn_cfg=dict(
type='CdnQueryGenerator',
noise_scale=dict(label=0.5, box=1.0),
group_cfg=dict(dynamic=True, num_groups=None, num_dn_queries=100)),
transformer=dict(
type='DinoTransformer',
two_stage_num_proposals=900,
encoder=dict(
type='DetrTransformerEncoder',
num_layers=6,
transformerlayers=dict(
type='BaseTransformerLayer',
attn_cfgs=dict(
type='MultiScaleDeformableAttention',
embed_dims=256,
dropout=0.0),
feedforward_channels=2048,
ffn_dropout=0.0, # 0.1 for DeformDETR
operation_order=('self_attn', 'norm', 'ffn', 'norm'))),
decoder=dict(
type='DinoTransformerDecoder',
num_layers=6,
return_intermediate=True,
transformerlayers=dict(
type='DetrTransformerDecoderLayer',
attn_cfgs=[
dict(
type='MultiheadAttention',
embed_dims=256,
num_heads=8,
dropout=0.0),
dict(
type='MultiScaleDeformableAttention',
embed_dims=256,
dropout=0.0),
],
feedforward_channels=2048,
ffn_dropout=0.0,
operation_order=('self_attn', 'norm', 'cross_attn', 'norm',
'ffn', 'norm')))),
positional_encoding=dict(
type='SinePositionalEncoding',
num_feats=128,
temperature=20,
normalize=True),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_bbox=dict(type='L1Loss', loss_weight=5.0),
loss_iou=dict(type='GIoULoss', loss_weight=2.0)),
# training and testing settings
train_cfg=dict(
assigner=dict(
type='HungarianAssigner',
cls_cost=dict(type='FocalLossCost', weight=2.0),
reg_cost=dict(type='BBoxL1Cost', weight=5.0, box_format='xywh'),
iou_cost=dict(type='IoUCost', iou_mode='giou', weight=2.0))),
test_cfg=dict(max_per_img=300))
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
# train_pipeline, NOTE the img_scale and the Pad's size_divisor is different
# from the default setting in mmdet.
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations', with_bbox=True),
dict(type='RandomFlip', flip_ratio=0.5),
dict(
type='AutoAugment',
policies=[
[
dict(
type='Resize',
img_scale=[(480, 1333), (512, 1333), (544, 1333),
(576, 1333), (608, 1333), (640, 1333),
(672, 1333), (704, 1333), (736, 1333),
(768, 1333), (800, 1333)],
multiscale_mode='value',
keep_ratio=True)
],
[
dict(
type='Resize',
img_scale=[(400, 4200), (500, 4200), (600, 4200)],
multiscale_mode='value',
keep_ratio=True),
dict(
type='RandomCrop',
crop_type='absolute_range',
crop_size=(384, 600),
allow_negative_crop=False),
dict(
type='Resize',
img_scale=[(480, 1333), (512, 1333), (544, 1333),
(576, 1333), (608, 1333), (640, 1333),
(672, 1333), (704, 1333), (736, 1333),
(768, 1333), (800, 1333)],
multiscale_mode='value',
override=True,
keep_ratio=True)
]
]),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels'])
]
# By default, models are trained on 8 GPUs with 4 images per GPU
data = dict(
samples_per_gpu=4,
train=dict(pipeline=train_pipeline))
# optimizer
optimizer = dict(
_delete_=True, type='AdamW', lr=0.0002, weight_decay=0.0001,
constructor='CustomLayerDecayOptimizerConstructor',
paramwise_cfg=dict(num_layers=30, layer_decay_rate=0.9,
depths=[4, 4, 18, 4]))
optimizer_config = dict(_delete_=True, grad_clip=dict(max_norm=0.1, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[11])
evaluation = dict(save_best='auto')
checkpoint_config = dict(
interval=1,
max_keep_ckpts=1,
save_last=True,
)
# fp16 = dict(loss_scale=512.)