256 lines
9.1 KiB
Python
Executable File
256 lines
9.1 KiB
Python
Executable File
#copyright (c) 2020 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.
|
|
|
|
from __future__ import absolute_import
|
|
from __future__ import division
|
|
from __future__ import print_function
|
|
|
|
import paddle.fluid as fluid
|
|
from paddle.fluid.initializer import MSRA
|
|
from paddle.fluid.param_attr import ParamAttr
|
|
|
|
__all__ = [
|
|
'MobileNetV3', 'MobileNetV3_small_x0_35', 'MobileNetV3_small_x0_5',
|
|
'MobileNetV3_small_x0_75', 'MobileNetV3_small_x1_0',
|
|
'MobileNetV3_small_x1_25', 'MobileNetV3_large_x0_35',
|
|
'MobileNetV3_large_x0_5', 'MobileNetV3_large_x0_75',
|
|
'MobileNetV3_large_x1_0', 'MobileNetV3_large_x1_25'
|
|
]
|
|
|
|
|
|
class MobileNetV3():
|
|
def __init__(self, params):
|
|
self.scale = params['scale']
|
|
model_name = params['model_name']
|
|
self.inplanes = 16
|
|
if model_name == "large":
|
|
self.cfg = [
|
|
# k, exp, c, se, nl, s,
|
|
[3, 16, 16, False, 'relu', 1],
|
|
[3, 64, 24, False, 'relu', (2, 1)],
|
|
[3, 72, 24, False, 'relu', 1],
|
|
[5, 72, 40, True, 'relu', (2, 1)],
|
|
[5, 120, 40, True, 'relu', 1],
|
|
[5, 120, 40, True, 'relu', 1],
|
|
[3, 240, 80, False, 'hard_swish', 1],
|
|
[3, 200, 80, False, 'hard_swish', 1],
|
|
[3, 184, 80, False, 'hard_swish', 1],
|
|
[3, 184, 80, False, 'hard_swish', 1],
|
|
[3, 480, 112, True, 'hard_swish', 1],
|
|
[3, 672, 112, True, 'hard_swish', 1],
|
|
[5, 672, 160, True, 'hard_swish', (2, 1)],
|
|
[5, 960, 160, True, 'hard_swish', 1],
|
|
[5, 960, 160, True, 'hard_swish', 1],
|
|
]
|
|
self.cls_ch_squeeze = 960
|
|
self.cls_ch_expand = 1280
|
|
elif model_name == "small":
|
|
self.cfg = [
|
|
# k, exp, c, se, nl, s,
|
|
[3, 16, 16, True, 'relu', (2, 1)],
|
|
[3, 72, 24, False, 'relu', (2, 1)],
|
|
[3, 88, 24, False, 'relu', 1],
|
|
[5, 96, 40, True, 'hard_swish', (2, 1)],
|
|
[5, 240, 40, True, 'hard_swish', 1],
|
|
[5, 240, 40, True, 'hard_swish', 1],
|
|
[5, 120, 48, True, 'hard_swish', 1],
|
|
[5, 144, 48, True, 'hard_swish', 1],
|
|
[5, 288, 96, True, 'hard_swish', (2, 1)],
|
|
[5, 576, 96, True, 'hard_swish', 1],
|
|
[5, 576, 96, True, 'hard_swish', 1],
|
|
]
|
|
self.cls_ch_squeeze = 576
|
|
self.cls_ch_expand = 1280
|
|
else:
|
|
raise NotImplementedError("mode[" + model_name +
|
|
"_model] is not implemented!")
|
|
|
|
supported_scale = [0.35, 0.5, 0.75, 1.0, 1.25]
|
|
assert self.scale in supported_scale, \
|
|
"supported scale are {} but input scale is {}".format(supported_scale, scale)
|
|
|
|
def __call__(self, input):
|
|
scale = self.scale
|
|
inplanes = self.inplanes
|
|
cfg = self.cfg
|
|
cls_ch_squeeze = self.cls_ch_squeeze
|
|
cls_ch_expand = self.cls_ch_expand
|
|
#conv1
|
|
conv = self.conv_bn_layer(
|
|
input,
|
|
filter_size=3,
|
|
num_filters=self.make_divisible(inplanes * scale),
|
|
stride=2,
|
|
padding=1,
|
|
num_groups=1,
|
|
if_act=True,
|
|
act='hard_swish',
|
|
name='conv1')
|
|
i = 0
|
|
inplanes = self.make_divisible(inplanes * scale)
|
|
for layer_cfg in cfg:
|
|
conv = self.residual_unit(
|
|
input=conv,
|
|
num_in_filter=inplanes,
|
|
num_mid_filter=self.make_divisible(scale * layer_cfg[1]),
|
|
num_out_filter=self.make_divisible(scale * layer_cfg[2]),
|
|
act=layer_cfg[4],
|
|
stride=layer_cfg[5],
|
|
filter_size=layer_cfg[0],
|
|
use_se=layer_cfg[3],
|
|
name='conv' + str(i + 2))
|
|
inplanes = self.make_divisible(scale * layer_cfg[2])
|
|
i += 1
|
|
|
|
conv = self.conv_bn_layer(
|
|
input=conv,
|
|
filter_size=1,
|
|
num_filters=self.make_divisible(scale * cls_ch_squeeze),
|
|
stride=1,
|
|
padding=0,
|
|
num_groups=1,
|
|
if_act=True,
|
|
act='hard_swish',
|
|
name='conv_last')
|
|
|
|
conv = fluid.layers.pool2d(
|
|
input=conv,
|
|
pool_size=2,
|
|
pool_stride=2,
|
|
pool_padding=0,
|
|
pool_type='max')
|
|
return conv
|
|
|
|
def conv_bn_layer(self,
|
|
input,
|
|
filter_size,
|
|
num_filters,
|
|
stride,
|
|
padding,
|
|
num_groups=1,
|
|
if_act=True,
|
|
act=None,
|
|
name=None,
|
|
use_cudnn=True,
|
|
res_last_bn_init=False):
|
|
conv = fluid.layers.conv2d(
|
|
input=input,
|
|
num_filters=num_filters,
|
|
filter_size=filter_size,
|
|
stride=stride,
|
|
padding=padding,
|
|
groups=num_groups,
|
|
act=None,
|
|
use_cudnn=use_cudnn,
|
|
param_attr=ParamAttr(name=name + '_weights'),
|
|
bias_attr=False)
|
|
bn_name = name + '_bn'
|
|
bn = fluid.layers.batch_norm(
|
|
input=conv,
|
|
param_attr=ParamAttr(
|
|
name=bn_name + "_scale",
|
|
regularizer=fluid.regularizer.L2DecayRegularizer(
|
|
regularization_coeff=0.0)),
|
|
bias_attr=ParamAttr(
|
|
name=bn_name + "_offset",
|
|
regularizer=fluid.regularizer.L2DecayRegularizer(
|
|
regularization_coeff=0.0)),
|
|
moving_mean_name=bn_name + '_mean',
|
|
moving_variance_name=bn_name + '_variance')
|
|
if if_act:
|
|
if act == 'relu':
|
|
bn = fluid.layers.relu(bn)
|
|
elif act == 'hard_swish':
|
|
bn = fluid.layers.hard_swish(bn)
|
|
return bn
|
|
|
|
def make_divisible(self, v, divisor=8, min_value=None):
|
|
if min_value is None:
|
|
min_value = divisor
|
|
new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
|
|
if new_v < 0.9 * v:
|
|
new_v += divisor
|
|
return new_v
|
|
|
|
def se_block(self, input, num_out_filter, ratio=4, name=None):
|
|
num_mid_filter = num_out_filter // ratio
|
|
pool = fluid.layers.pool2d(
|
|
input=input, pool_type='avg', global_pooling=True, use_cudnn=False)
|
|
conv1 = fluid.layers.conv2d(
|
|
input=pool,
|
|
filter_size=1,
|
|
num_filters=num_mid_filter,
|
|
act='relu',
|
|
param_attr=ParamAttr(name=name + '_1_weights'),
|
|
bias_attr=ParamAttr(name=name + '_1_offset'))
|
|
conv2 = fluid.layers.conv2d(
|
|
input=conv1,
|
|
filter_size=1,
|
|
num_filters=num_out_filter,
|
|
act='hard_sigmoid',
|
|
param_attr=ParamAttr(name=name + '_2_weights'),
|
|
bias_attr=ParamAttr(name=name + '_2_offset'))
|
|
scale = fluid.layers.elementwise_mul(x=input, y=conv2, axis=0)
|
|
return scale
|
|
|
|
def residual_unit(self,
|
|
input,
|
|
num_in_filter,
|
|
num_mid_filter,
|
|
num_out_filter,
|
|
stride,
|
|
filter_size,
|
|
act=None,
|
|
use_se=False,
|
|
name=None):
|
|
|
|
conv0 = self.conv_bn_layer(
|
|
input=input,
|
|
filter_size=1,
|
|
num_filters=num_mid_filter,
|
|
stride=1,
|
|
padding=0,
|
|
if_act=True,
|
|
act=act,
|
|
name=name + '_expand')
|
|
|
|
conv1 = self.conv_bn_layer(
|
|
input=conv0,
|
|
filter_size=filter_size,
|
|
num_filters=num_mid_filter,
|
|
stride=stride,
|
|
padding=int((filter_size - 1) // 2),
|
|
if_act=True,
|
|
act=act,
|
|
num_groups=num_mid_filter,
|
|
use_cudnn=False,
|
|
name=name + '_depthwise')
|
|
if use_se:
|
|
conv1 = self.se_block(
|
|
input=conv1, num_out_filter=num_mid_filter, name=name + '_se')
|
|
|
|
conv2 = self.conv_bn_layer(
|
|
input=conv1,
|
|
filter_size=1,
|
|
num_filters=num_out_filter,
|
|
stride=1,
|
|
padding=0,
|
|
if_act=False,
|
|
name=name + '_linear',
|
|
res_last_bn_init=True)
|
|
if num_in_filter != num_out_filter or stride != 1:
|
|
return conv2
|
|
else:
|
|
return fluid.layers.elementwise_add(x=input, y=conv2, act=None)
|