-
Notifications
You must be signed in to change notification settings - Fork 2
/
convnext_v2.py
294 lines (196 loc) · 7.63 KB
/
convnext_v2.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
288
289
290
291
292
293
294
# ================================================================
# MIT License
# Copyright (c) 2023 edwardyehuang (https://github.com/edwardyehuang)
# ================================================================
# This code implemented https://arxiv.org/pdf/2301.00808.pdf
import tensorflow as tf
import numpy as np
from iseg.utils.drops import drop_path
from iseg.backbones.utils.layerwise_decay import decay_layers_lr
from iseg.utils.keras3_utils import Keras3_Layer_Wrapper, Keras3_Model_Wrapper
class GlobalResponseNormlizationLayer(Keras3_Layer_Wrapper):
def __init__(self, trainable=True, epsilon=1e-6, name=None):
super().__init__(trainable=trainable, name=name)
self.epsilon = epsilon
def build(self, input_shape):
channels = input_shape[-1]
self.gamma = self.add_weight(
name="gamma",
shape=[1, 1, 1, channels],
trainable=self.trainable,
initializer=tf.zeros_initializer(),
dtype=tf.float32,
)
self.beta = self.add_weight(
name="beta",
shape=[1, 1, 1, channels],
trainable=self.trainable,
initializer=tf.zeros_initializer(),
dtype=tf.float32,
)
super().build(input_shape)
def call (self, inputs, training=None):
x = inputs
x = tf.cast(x, tf.float32)
beta = tf.cast(self.beta, tf.float32)
gamma = tf.cast(self.gamma, tf.float32)
gx = tf.pow((tf.reduce_sum(tf.pow(x, 2), axis=(1, 2), keepdims=True) + self.epsilon), 0.5)
# gx = tf.nn.l2_normalize(x, axis=(1, 2), epsilon=1e-6)
nx = gx / (tf.reduce_mean(gx, axis=-1, keepdims=True) + self.epsilon)
x = gamma * (x * nx) + beta + x
return tf.cast(x, inputs.dtype)
class Block(Keras3_Model_Wrapper):
def __init__(self, filters, drop_path_prob=0.0, name=None):
super().__init__(name=name)
self.drop_path_prob = drop_path_prob
self.filters = filters
self.dwconv = tf.keras.layers.DepthwiseConv2D(kernel_size=7, padding="same", name=f"{self.name}/dwconv")
self.norm = tf.keras.layers.LayerNormalization(epsilon=1e-6, name=f"{self.name}/norm")
self.pwconv1 = tf.keras.layers.Dense(units=4 * filters, name=f"{self.name}/pwconv1")
self.grn = GlobalResponseNormlizationLayer(trainable=self.trainable, name=f"{self.name}/grn")
self.pwconv2 = tf.keras.layers.Dense(units=filters, name=f"{self.name}/pwconv2")
def build(self, input_shape):
super().build(input_shape)
def call(self, inputs, training=None):
x = inputs
x = self.dwconv(x)
x = self.norm(x)
x = self.pwconv1(x)
x = tf.nn.gelu(x)
x = self.grn(x)
x = self.pwconv2(x)
if self.drop_path_prob != 0.0 and training:
x = drop_path(x, drop_prob=self.drop_path_prob, training=training)
x += tf.cast(inputs, x.dtype)
return x
class DownSampleLayer(Keras3_Model_Wrapper):
def __init__(self, filters=96, strides=2, swap=False, name=None):
super().__init__(name=name)
self.swap = swap
names = ["1", "0"] if swap else ["0", "1"]
self.norm = tf.keras.layers.LayerNormalization(epsilon=1e-6, name=f"{self.name}/{names[0]}")
self.conv = tf.keras.layers.Conv2D(
filters=filters, kernel_size=strides, strides=strides, padding="same", name=f"{self.name}/{names[1]}"
)
def build(self, input_shape):
super().build(input_shape)
def call(self, inputs, training=None):
x = inputs
if self.swap:
x = self.norm(self.conv(x))
else:
x = self.conv(self.norm(x))
return x
class Stage(Keras3_Model_Wrapper):
def __init__(self, filters=96, depth=3, drop_path_probs=[], name=None):
super().__init__(name=name)
assert len(drop_path_probs) == 0 or len(drop_path_probs) == depth
self.blocks = [
Block(
filters=filters,
drop_path_prob=drop_path_probs[i],
name=f"{self.name}/{i}",
)
for i in range(depth)
]
def build(self, input_shape):
super().build(input_shape)
def call(self, inputs, training=None):
x = inputs
for block in self.blocks:
x = block(x, training=training)
return x
class ConvNeXtV2(Keras3_Model_Wrapper):
def __init__(
self,
depths=[3, 3, 9, 3],
filters_list=[96, 192, 384, 768],
drop_path_rate=0.0,
return_endpoints=False,
name=None,
):
super().__init__(name=name)
self.return_endpoints = return_endpoints
num_stage = len(depths)
assert num_stage == len(filters_list)
drop_path_rates = np.linspace(0.0, drop_path_rate, sum(depths))
self.downsample_blocks = []
self.stages = []
cur = 0
for i in range(num_stage):
self.downsample_blocks += [
DownSampleLayer(
filters=filters_list[i], strides=4 if i == 0 else 2, swap=i == 0, name=f"downsample_layers/{i}"
)
]
self.stages += [
Stage(
filters=filters_list[i],
depth=depths[i],
drop_path_probs=drop_path_rates[cur : cur + depths[i]],
name=f"stages/{i}",
)
]
cur += depths[i]
def build(self, input_shape):
super().build(input_shape)
def call(self, inputs, training=None):
x = inputs
endpoints = [None]
for i in range(len(self.stages)):
x = self.downsample_blocks[i](x, training=training)
x = self.stages[i](x, training=training)
endpoints += [x]
if self.return_endpoints:
return endpoints
return x
def decay_lr(
self,
rate=0.99
):
stages = list(self.stages)
stages.reverse()
decay_layers_lr(stages, rate=rate)
def convnext_v2_nano(return_endpoints=False):
return ConvNeXtV2(
depths=[2, 2, 8, 2],
filters_list=[80, 160, 320, 640],
return_endpoints=return_endpoints,
drop_path_rate=0.1,
)
def convnext_v2_tiny(return_endpoints=False):
return ConvNeXtV2(
depths=[3, 3, 9, 3],
filters_list=[96, 192, 384, 768],
return_endpoints=return_endpoints,
drop_path_rate=0.1,
)
def convnext_v2_large(return_endpoints=False):
return ConvNeXtV2(
depths=[3, 3, 27, 3],
filters_list=[192, 384, 768, 1536],
return_endpoints=return_endpoints,
drop_path_rate=0.3,
)
def convnext_v2_huge(return_endpoints=False):
return ConvNeXtV2(
depths=[3, 3, 27, 3],
filters_list=[352, 704, 1408, 2816],
return_endpoints=return_endpoints,
drop_path_rate=0.4,
)
def build_dilated_convnext(model: ConvNeXtV2, output_stride=32):
num_stages = len(model.stages)
current_os = 1
current_dilation = 1
for i in range(num_stages):
if current_os >= output_stride:
current_dilation *= model.downsample_blocks[i].conv.strides[0]
model.downsample_blocks[i].conv.strides = (1, 1)
model.downsample_blocks[i].conv.dilation_rate = (current_dilation, current_dilation)
for block in model.stages[i].blocks:
block.dwconv.strides = (1, 1)
block.dwconv.dilation_rate = (current_dilation, current_dilation)
else:
current_os *= model.downsample_blocks[i].conv.strides[0]
return model