add mobilenetv2
louzan
parent
6515bb999c
commit
ba391c029a
|
|
@ -0,0 +1,275 @@
|
|||
import logging
|
||||
|
||||
import torch.nn as nn
|
||||
import torch.utils.checkpoint as cp
|
||||
|
||||
from ..runner import load_checkpoint
|
||||
from .base_backbone import BaseBackbone
|
||||
from .weight_init import constant_init, kaiming_init
|
||||
|
||||
|
||||
def conv3x3(in_planes, out_planes, stride=1, dilation=1):
|
||||
"""3x3 convolution with padding"""
|
||||
return nn.Conv2d(
|
||||
in_planes,
|
||||
out_planes,
|
||||
kernel_size=3,
|
||||
stride=stride,
|
||||
padding=dilation,
|
||||
dilation=dilation,
|
||||
bias=False)
|
||||
|
||||
|
||||
def conv_1x1_bn(inp, oup, act=nn.ReLU6):
|
||||
return nn.Sequential(
|
||||
nn.Conv2d(inp, oup, 1, 1, 0, bias=False),
|
||||
nn.BatchNorm2d(oup),
|
||||
act(inplace=True)
|
||||
)
|
||||
|
||||
|
||||
class ConvBNReLU(nn.Sequential):
|
||||
def __init__(self,
|
||||
in_planes,
|
||||
out_planes,
|
||||
kernel_size=3,
|
||||
stride=1,
|
||||
groups=1,
|
||||
activation=nn.ReLU6):
|
||||
padding = (kernel_size - 1) // 2
|
||||
|
||||
try:
|
||||
self.activation = activation(inplace=True)
|
||||
except RuntimeWarning('inplace is not allowed to use'):
|
||||
self.activation = activation()
|
||||
|
||||
super(ConvBNReLU, self).__init__(
|
||||
nn.Conv2d(in_planes,
|
||||
out_planes,
|
||||
kernel_size,
|
||||
stride,
|
||||
padding,
|
||||
groups=groups,
|
||||
bias=False),
|
||||
nn.BatchNorm2d(out_planes),
|
||||
self.activation
|
||||
)
|
||||
|
||||
|
||||
def _make_divisible(v, divisor, min_value=None):
|
||||
if min_value is None:
|
||||
min_value = divisor
|
||||
new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
|
||||
# Make sure that round down does not go down by more than 10%.
|
||||
if new_v < 0.9 * v:
|
||||
new_v += divisor
|
||||
return new_v
|
||||
|
||||
|
||||
class InvertedResidual(nn.Module):
|
||||
def __init__(self,
|
||||
inplanes,
|
||||
outplanes,
|
||||
stride,
|
||||
expand_ratio,
|
||||
activation=nn.ReLU6,
|
||||
with_cp=False):
|
||||
super(InvertedResidual, self).__init__()
|
||||
self.stride = stride
|
||||
assert stride in [1, 2]
|
||||
self.with_cp = with_cp
|
||||
self.use_res_connect = self.stride == 1 and inplanes == outplanes
|
||||
hidden_dim = int(round(inplanes * expand_ratio))
|
||||
|
||||
layers = []
|
||||
if expand_ratio != 1:
|
||||
# pw
|
||||
layers.append(ConvBNReLU(inplanes,
|
||||
hidden_dim,
|
||||
kernel_size=1,
|
||||
activation=activation))
|
||||
layers.extend([
|
||||
# dw
|
||||
ConvBNReLU(hidden_dim,
|
||||
hidden_dim,
|
||||
stride=stride,
|
||||
groups=hidden_dim,
|
||||
activation=activation),
|
||||
# pw-linear
|
||||
nn.Conv2d(hidden_dim, outplanes, 1, 1, 0, bias=False),
|
||||
nn.BatchNorm2d(outplanes),
|
||||
])
|
||||
self.conv = nn.Sequential(*layers)
|
||||
|
||||
def forward(self, x):
|
||||
def _inner_forward(x):
|
||||
if self.use_res_connect:
|
||||
return x + self.conv(x)
|
||||
else:
|
||||
return self.conv(x)
|
||||
|
||||
if self.with_cp and x.requires_grad:
|
||||
out = cp.checkpoint(_inner_forward, x)
|
||||
else:
|
||||
out = _inner_forward(x)
|
||||
|
||||
return out
|
||||
|
||||
|
||||
def make_inverted_res_layer(block,
|
||||
inplanes,
|
||||
planes,
|
||||
num_blocks,
|
||||
stride=1,
|
||||
expand_ratio=6,
|
||||
activation_type=nn.ReLU6,
|
||||
with_cp=False):
|
||||
layers = []
|
||||
for i in range(num_blocks):
|
||||
if i == 0:
|
||||
layers.append(block(inplanes, planes, stride,
|
||||
expand_ratio=expand_ratio,
|
||||
activation=activation_type,
|
||||
with_cp=with_cp))
|
||||
else:
|
||||
layers.append(block(inplanes, planes, 1,
|
||||
expand_ratio=expand_ratio,
|
||||
activation=activation_type,
|
||||
with_cp=with_cp))
|
||||
return nn.Sequential(*layers)
|
||||
|
||||
|
||||
class MobileNetv2(BaseBackbone):
|
||||
"""MobileNetv2 backbone.
|
||||
|
||||
Args:
|
||||
widen_factor (float): Config of widen_factor.
|
||||
activation (str): Activation type of the network.
|
||||
out_indices (Sequence[int]): Output from which stages.
|
||||
frozen_stages (int): Stages to be frozen (all param fixed). -1 means
|
||||
not freezing any parameters.
|
||||
bn_eval (bool): Whether to set BN layers as eval mode, namely, freeze
|
||||
running stats (mean and var).
|
||||
bn_frozen (bool): Whether to freeze weight and bias of BN layers.
|
||||
with_cp (bool): Use checkpoint or not. Using checkpoint will save some
|
||||
memory while slowing down the training speed.
|
||||
"""
|
||||
|
||||
def __init__(self,
|
||||
widen_factor=1.,
|
||||
activation=nn.ReLU6,
|
||||
out_indices=(0, 1, 2, 3, 4, 5, 6),
|
||||
frozen_stages=-1,
|
||||
bn_eval=True,
|
||||
bn_frozen=False,
|
||||
with_cp=False):
|
||||
super(MobileNetv2, self).__init__()
|
||||
block = InvertedResidual
|
||||
|
||||
inverted_residual_setting = {
|
||||
# lager_index: [expand_ratio, out_channel, n, stide]
|
||||
0: [1, 16, 1, 1],
|
||||
1: [6, 24, 2, 2],
|
||||
2: [6, 32, 3, 2],
|
||||
3: [6, 64, 4, 2],
|
||||
4: [6, 96, 3, 1],
|
||||
5: [6, 160, 3, 2],
|
||||
6: [6, 320, 1, 1]
|
||||
}
|
||||
self.widen_factor = widen_factor
|
||||
self.activation_type = activation
|
||||
try:
|
||||
self.activation = activation(inplace=True)
|
||||
except RuntimeWarning('inplace is not allowed to use'):
|
||||
self.activation = activation()
|
||||
|
||||
self.out_indices = out_indices
|
||||
self.frozen_stages = frozen_stages
|
||||
self.bn_eval = bn_eval
|
||||
self.bn_frozen = bn_frozen
|
||||
self.with_cp = with_cp
|
||||
|
||||
self.inplanes = 32
|
||||
self.inplanes = _make_divisible(self.inplanes * widen_factor, 8)
|
||||
self.conv1 = conv3x3(3, self.inplanes, stride=2)
|
||||
|
||||
self.inverted_res_layers = []
|
||||
for i, later_cfg in enumerate(inverted_residual_setting):
|
||||
t, c, n, s = later_cfg
|
||||
planes = _make_divisible(c * widen_factor, 8)
|
||||
inverted_res_layer = make_inverted_res_layer(
|
||||
block,
|
||||
self.inplanes,
|
||||
planes,
|
||||
num_blocks=n,
|
||||
stride=s,
|
||||
expand_ratio=t,
|
||||
activation_type=self.activation_type,
|
||||
with_cp=self.with_cp)
|
||||
self.inplanes = planes
|
||||
layer_name = 'layer{}'.format(i + 1)
|
||||
self.add_module(layer_name, inverted_res_layer)
|
||||
self.inverted_res_layers.append(layer_name)
|
||||
|
||||
self.out_channel = 1280
|
||||
self.out_channel = int(self.out_channel * widen_factor) \
|
||||
if widen_factor > 1.0 else self.out_channel
|
||||
self.conv1_bn = conv_1x1_bn(self.inplanes, self.out_channel)
|
||||
|
||||
self.feat_dim = self.out_channel
|
||||
|
||||
def init_weights(self, pretrained=None):
|
||||
if isinstance(pretrained, str):
|
||||
logger = logging.getLogger()
|
||||
load_checkpoint(self, pretrained, strict=False, logger=logger)
|
||||
elif pretrained is None:
|
||||
for m in self.modules():
|
||||
if isinstance(m, nn.Conv2d):
|
||||
kaiming_init(m)
|
||||
elif isinstance(m, nn.BatchNorm2d):
|
||||
constant_init(m, 1)
|
||||
else:
|
||||
raise TypeError('pretrained must be a str or None')
|
||||
|
||||
def forward(self, x):
|
||||
x = self.conv1(x)
|
||||
x = self.bn1(x)
|
||||
x = self.activation(x)
|
||||
|
||||
outs = []
|
||||
for i, layer_name in enumerate(self.inverted_res_layers):
|
||||
inverted_res_layer = getattr(self, layer_name)
|
||||
x = inverted_res_layer(x)
|
||||
if i in self.out_indices:
|
||||
outs.append(x)
|
||||
|
||||
x = self.conv1_bn(x)
|
||||
outs.append(x)
|
||||
|
||||
if len(outs) == 1:
|
||||
return outs[0]
|
||||
else:
|
||||
return tuple(outs)
|
||||
|
||||
def train(self, mode=True):
|
||||
super(MobileNetv2, self).train(mode)
|
||||
if self.bn_eval:
|
||||
for m in self.modules():
|
||||
if isinstance(m, nn.BatchNorm2d):
|
||||
m.eval()
|
||||
if self.bn_frozen:
|
||||
for params in m.parameters():
|
||||
params.requires_grad = False
|
||||
if mode and self.frozen_stages >= 0:
|
||||
for param in self.conv1.parameters():
|
||||
param.requires_grad = False
|
||||
for param in self.bn1.parameters():
|
||||
param.requires_grad = False
|
||||
self.bn1.eval()
|
||||
self.bn1.weight.requires_grad = False
|
||||
self.bn1.bias.requires_grad = False
|
||||
for i in range(1, self.frozen_stages + 1):
|
||||
mod = getattr(self, 'layer{}'.format(i))
|
||||
mod.eval()
|
||||
for param in mod.parameters():
|
||||
param.requires_grad = False
|
||||
Loading…
Reference in New Issue