-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathmodel.py
155 lines (125 loc) · 3.95 KB
/
model.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
from src.nn.rtdetr.hybrid_encoder import HybridEncoder
from src.nn.rtdetr.rtdetr import RTDETR
from src.nn.rtdetr.rtdetr_decoder import RTDETRTransformer
from src.nn.backbone.presnet import PResNet
def r50vd_backbone(
depth=50,
variant='d',
freeze_at=0,
return_idx=[1, 2, 3],
num_stages=4,
freeze_norm=True,
pretrained=True):
return PResNet(
depth=depth,
variant=variant,
freeze_at=freeze_at,
return_idx=return_idx,
num_stages=num_stages,
freeze_norm=freeze_norm,
pretrained=pretrained)
def r50vd_encoder(
in_channels=[512, 1024, 2048],
feat_strides=[8, 16, 32],
hidden_dim=256,
use_encoder_idx=[2],
num_encoder_layers=1,
nhead=8,
dim_feedforward=1024,
dropout=0.0,
enc_act='gelu',
pe_temperature=10000,
expansion=1.0,
depth_mult=1,
act='silu',
eval_spatial_size=[640, 640]):
return HybridEncoder(
in_channels=in_channels,
feat_strides=feat_strides,
hidden_dim=hidden_dim,
use_encoder_idx=use_encoder_idx,
num_encoder_layers=num_encoder_layers,
nhead=nhead,
dim_feedforward=dim_feedforward,
dropout=dropout,
enc_act=enc_act,
pe_temperature=pe_temperature,
expansion=expansion,
depth_mult=depth_mult,
act=act,
eval_spatial_size=eval_spatial_size)
def r50vd_decoder(
feat_channels=[256, 256, 256],
feat_strides=[8, 16, 32],
hidden_dim=256,
num_levels=3,
num_queries=300,
num_decoder_layers=6,
num_denoising=100,
eval_idx=-1,
eval_spatial_size=[640, 640]):
return RTDETRTransformer(
feat_channels=feat_channels,
feat_strides=feat_strides,
hidden_dim=hidden_dim,
num_levels=num_levels,
num_queries=num_queries,
num_decoder_layers=num_decoder_layers,
num_denoising=num_denoising,
eval_idx=eval_idx,
eval_spatial_size=eval_spatial_size)
def r50vd():
backbone = r50vd_backbone()
encoder = r50vd_encoder()
decoder = r50vd_decoder()
model = RTDETR(
backbone=backbone,
encoder=encoder,
decoder=decoder,
multi_scale=[480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
)
return model
def r18vd():
backbone = r50vd_backbone(depth=18, freeze_at=-1, freeze_norm=False)
encoder = r50vd_encoder(in_channels=[128, 256, 512], expansion=0.5)
decoder = r50vd_decoder(num_decoder_layers=3)
model = RTDETR(
backbone=backbone,
encoder=encoder,
decoder=decoder,
multi_scale=[480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
)
return model
def r34vd():
backbone = r50vd_backbone(depth=34, freeze_at=-1, freeze_norm=False)
encoder = r50vd_encoder(in_channels=[128, 256, 512], expansion=0.5)
decoder = r50vd_decoder(num_decoder_layers=4)
model = RTDETR(
backbone=backbone,
encoder=encoder,
decoder=decoder,
multi_scale=[480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
)
return model
def r50vd_m():
backbone = r50vd_backbone()
encoder = r50vd_encoder(expansion=0.5)
decoder = r50vd_decoder(eval_idx=2)
model = RTDETR(
backbone=backbone,
encoder=encoder,
decoder=decoder,
multi_scale=[480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
)
return model
def r101vd():
backbone = r50vd_backbone(depth=101)
encoder = r50vd_encoder(hidden_dim=384, dim_feedforward=2048)
decoder = r50vd_decoder(feat_channels=[384, 384, 384])
model = RTDETR(
backbone=backbone,
encoder=encoder,
decoder=decoder,
multi_scale=[480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
)
return model