-
Notifications
You must be signed in to change notification settings - Fork 1.7k
/
Copy pathefficientformerv2_seg.py
287 lines (248 loc) · 10.8 KB
/
efficientformerv2_seg.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
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
# copyright (c) 2023 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import numpy as np
import paddle.nn as nn
import paddle.nn.functional as F
from paddleseg.cvlibs import manager, param_init
from paddleseg.utils import utils
@manager.MODELS.add_component
class EfficientFormerSeg(nn.Layer):
"""
The EfficientFormerV2 implementation based on PaddlePaddle.
The original article refers to Yanyu Li, Ju Hu, Yang Wen, Georgios Evangelidis,
Kamyar Salahi, Yanzhi Wang, Sergey Tulyakov, Jian Ren.
"Rethinking Vision Transformers for MobileNet Size and Speed".
(https://arxiv.org/pdf/2212.08059.pdf).
Args:
backbone (paddle.nn.Layer): Backbone networks.
num_classes (int): The unique number of target classes.
backbone_indices (list[int], optional): The values in the tuple indicate the indices of output of backbone.
Default: [0, 1, 2, 3].
align_corners (bool, optional): An argument of F.interpolate. It should be set to False when the output size of feature
is even, e.g. 1024x512, otherwise it is True, e.g. 769x769. Default: False.
head_channels (int, optional): The number of hidden channels of segmentation head. Default: 128.
add_extra_convs (bool|str): An argument about whether to add extra conv or not or where to add extra conv. Default: False.
pretrained (str, optional): The path or url of pretrained model. Default: None.
"""
def __init__(self,
backbone,
num_classes,
backbone_indices=[0, 1, 2, 3],
align_corners=False,
head_channels=128,
add_extra_convs=False,
pretrained=None):
super().__init__()
self.align_corners = align_corners
self.backbone = backbone
self.pretrained = pretrained
backbone_channels = [
backbone.feat_channels[i] for i in backbone_indices
]
self.neck = EfficientFormerFPNNeck(
in_channels=backbone_channels,
out_channels=256,
num_outs=4,
add_extra_convs=add_extra_convs)
self.head = EfficientFormerFPN(
in_channels=[256, 256, 256, 256],
in_index=[0, 1, 2, 3],
feature_strides=[4, 8, 16, 32],
channels=head_channels,
dropout_ratio=0.1,
num_classes=num_classes,
align_corners=self.align_corners)
self.init_weight()
def forward(self, x):
H, W = x.shape[-2:]
x = self.backbone(x)
x = self.neck(x)
x = self.head(x)
x = [
F.interpolate(
x,
size=[H, W],
mode='bilinear',
align_corners=self.align_corners)
]
return x
def init_weight(self):
if self.pretrained is not None:
utils.load_entire_model(self, self.pretrained)
class EfficientFormerFPNNeck(nn.Layer):
def __init__(self,
in_channels,
out_channels,
num_outs,
start_level=0,
end_level=-1,
add_extra_convs=False,
extra_convs_on_inputs=False,
relu_before_extra_convs=False):
super().__init__()
assert isinstance(in_channels, list)
self.in_channels = in_channels
self.out_channels = out_channels
self.num_ins = len(in_channels)
self.num_outs = num_outs
self.relu_before_extra_convs = relu_before_extra_convs
if end_level == -1:
self.backbone_end_level = self.num_ins
assert num_outs >= self.num_ins - start_level
else:
self.backbone_end_level = end_level
assert end_level <= len(in_channels)
assert num_outs == end_level - start_level
self.start_level = start_level
self.end_level = end_level
self.add_extra_convs = add_extra_convs
assert isinstance(add_extra_convs, (str, bool))
if isinstance(add_extra_convs, str):
assert add_extra_convs in ('on_input', 'on_lateral', 'on_output')
elif add_extra_convs: # True
if extra_convs_on_inputs:
self.add_extra_convs = 'on_input'
else:
self.add_extra_convs = 'on_output'
self.lateral_convs = nn.LayerList()
self.fpn_convs = nn.LayerList()
for i in range(self.start_level, self.backbone_end_level):
l_conv = nn.Conv2D(in_channels[i], out_channels, 1)
fpn_conv = nn.Conv2D(out_channels, out_channels, 3, padding=1)
self.lateral_convs.append(l_conv)
self.fpn_convs.append(fpn_conv)
# add extra conv layers (e.g., RetinaNet)
extra_levels = num_outs - self.backbone_end_level + self.start_level
if self.add_extra_convs and extra_levels >= 1:
for i in range(extra_levels):
if i == 0 and self.add_extra_convs == 'on_input':
in_channels = self.in_channels[self.backbone_end_level - 1]
else:
in_channels = out_channels
extra_fpn_conv = nn.Conv2D(
in_channels, out_channels, 3, stride=2, padding=1)
self.fpn_convs.append(extra_fpn_conv)
self.init_weight()
def forward(self, inputs):
assert len(inputs) == len(self.in_channels)
laterals = [
lateral_conv(inputs[i + self.start_level])
for i, lateral_conv in enumerate(self.lateral_convs)
]
used_backbone_levels = len(laterals)
for i in range(used_backbone_levels - 1, 0, -1):
prev_shape = laterals[i - 1].shape[2:]
laterals[i - 1] += F.interpolate(
laterals[i],
size=prev_shape,
mode='nearest',
align_corners=False)
outs = [
self.fpn_convs[i](laterals[i]) for i in range(used_backbone_levels)
]
if self.num_outs > len(outs):
if not self.add_extra_convs:
for i in range(self.num_outs - used_backbone_levels):
outs.append(F.max_pool2d(outs[-1], 1, stride=2))
else:
if self.add_extra_convs == 'on_input':
extra_source = inputs[self.backbone_end_level - 1]
elif self.add_extra_convs == 'on_lateral':
extra_source = laterals[-1]
elif self.add_extra_convs == 'on_output':
extra_source = outs[-1]
else:
raise NotImplementedError
outs.append(self.fpn_convs[used_backbone_levels](extra_source))
for i in range(used_backbone_levels + 1, self.num_outs):
if self.relu_before_extra_convs:
outs.append(self.fpn_convs[i](F.relu(outs[-1])))
else:
outs.append(self.fpn_convs[i](outs[-1]))
return tuple(outs)
def init_weight(self):
for sublayer in self.sublayers():
if isinstance(sublayer, nn.Conv2D):
param_init.kaiming_normal_init(sublayer.weight)
elif isinstance(sublayer, (nn.BatchNorm, nn.SyncBatchNorm)):
param_init.constant_init(sublayer.weight, value=1.0)
param_init.constant_init(sublayer.bias, value=0.0)
class EfficientFormerFPN(nn.Layer):
def __init__(self,
in_index=[0, 1, 2, 3],
in_channels=[256, 256, 256, 256],
channels=128,
feature_strides=[4, 8, 16, 32],
dropout_ratio=0.1,
num_classes=150,
align_corners=False):
super().__init__()
self.in_channels = in_channels
assert len(feature_strides) == len(self.in_channels)
assert min(feature_strides) == feature_strides[0]
self.in_index = in_index
self.channels = channels
self.feature_strides = feature_strides
self.align_corners = align_corners
self.dropout_ratio = dropout_ratio
self.num_classes = num_classes
if self.dropout_ratio > 0:
self.dropout = nn.Dropout2D(self.dropout_ratio)
self.scale_heads = nn.LayerList()
for i in range(len(feature_strides)):
head_length = max(
1,
int(np.log2(feature_strides[i]) - np.log2(feature_strides[0])))
scale_head = []
for k in range(head_length):
scale_head.append(
nn.Sequential(
nn.Conv2D(
self.in_channels[i] if k == 0 else self.channels,
self.channels,
3,
padding=1,
bias_attr=False),
nn.BatchNorm2D(self.channels),
nn.ReLU()))
if feature_strides[i] != feature_strides[0]:
scale_head.append(
nn.Upsample(
scale_factor=2,
mode='bilinear',
align_corners=self.align_corners))
self.scale_heads.append(nn.Sequential(*scale_head))
self.cls_seg = nn.Conv2D(self.channels, self.num_classes, kernel_size=1)
self.init_weight()
def forward(self, inputs):
x = [inputs[i] for i in self.in_index]
output = self.scale_heads[0](x[0])
for i in range(1, len(self.feature_strides)):
# non inplace
output = output + F.interpolate(
self.scale_heads[i](x[i]),
size=output.shape[2:],
mode='bilinear',
align_corners=self.align_corners)
if self.dropout_ratio > 0:
output = self.dropout(output)
output = self.cls_seg(output)
return output
def init_weight(self):
for sublayer in self.sublayers():
if isinstance(sublayer, nn.Conv2D):
param_init.kaiming_normal_init(sublayer.weight)
elif isinstance(sublayer, (nn.BatchNorm, nn.SyncBatchNorm)):
param_init.constant_init(sublayer.weight, value=1.0)
param_init.constant_init(sublayer.bias, value=0.0)