GhostNet代码理解
class GhostNet(nn.Module):
def __init__(self, cfgs, num_classes=1000, width_mult=1.):
super(GhostNet, self).__init__()
# setting of inverted residual blocks
if cfgs is None:
cfgs = [
# k, t, c, SE, s
[3, 16, 16, 0, 1],
[3, 48, 24, 0, 2],
[3, 72, 24, 0, 1],
[5, 72, 40, 1, 2],
[5, 120, 40, 1, 1],
[3, 240, 80, 0, 2],
[3, 200, 80, 0, 1],
[3, 184, 80, 0, 1],
[3, 184, 80, 0, 1],
[3, 480, 112, 1, 1],
[3, 672, 112, 1, 1],
[5, 672, 160, 1, 2],
[5, 960, 160, 0, 1],
[5, 960, 160, 1, 1],
[5, 960, 160, 0, 1],
[5, 960, 160, 1, 1]
]
self.cfgs = cfgs
# building first layer
output_channel = _make_divisible(16 * width_mult, 4)
layers = [nn.Sequential(
nn.Conv2d(3, output_channel, 3, 2, 1, bias=False),
nn.BatchNorm2d(output_channel),
nn.ReLU(inplace=True)
)]
input_channel = output_channel
# building inverted residual blocks
block = GhostBottleneck
for k, exp_size, c, use_se, s in self.cfgs: # 遍历参数
output_channel = _make_divisible(c * width_mult, 4)
hidden_channel = _make_divisible(exp_size * width_mult, 4)
layers.append(block(input_channel, hidden_channel, output_channel, k, s, use_se)) # 追加GhostBottleneck模块
input_channel = output_channel # 更新输入通道数
self.features = nn.Sequential(*layers)
# building last several layers
output_channel = _make_divisible(exp_size * width_mult, 4)
self.squeeze = nn.Sequential(
nn.Conv2d(input_channel, output_channel, 1, 1, 0, bias=False),
nn.BatchNorm2d(output_channel),
nn.ReLU(inplace=True),
nn.AdaptiveAvgPool2d((1, 1)),
)
input_channel = output_channel # 更新输入通道数
output_channel = 1280
self.classifier = nn.Sequential(
nn.Linear(input_channel, output_channel, bias=False),
nn.BatchNorm1d(output_channel),
nn.ReLU(inplace=True),
nn.Linear(output_channel, num_classes),
)
self._initialize_weights()
def forward(self, x):
x = self.features(x)
x = self.squeeze(x)
x = x.view(x.size(0), -1)
x = self.classifier(x)
return x
def _initialize_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def _make_divisible(v, divisor, min_value=None):
"""
This function is taken from the original tf repo.
It ensures that all layers have a channel number that is divisible by 8
It can be seen here:
https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet/mobilenet.py
"""
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 GhostBottleneck(nn.Module):
def __init__(self, inp, hidden_dim, oup, kernel_size, stride, use_se):
super(GhostBottleneck, self).__init__()
assert stride in [1, 2]
self.conv = nn.Sequential(
# pw # 第一个GhostModule
GhostModule(inp, hidden_dim, kernel_size=1, relu=True),
# dw 如果步长为2，用可分离卷积，否则不用
depthwise_conv(hidden_dim, hidden_dim, kernel_size, stride, relu=False) if stride == 2 else nn.Sequential(),
# Squeeze-and-Excite
SELayer(hidden_dim) if use_se else nn.Sequential(), # se模块选择
# pw-linear 第二个GhostModule
GhostModule(hidden_dim, oup, kernel_size=1, relu=False),
)
if stride == 1 and inp == oup: #
self.shortcut = nn.Sequential() # 为空
else: # 步长不为1 或者输入与输出通道数不等，则用shortcut操作
self.shortcut = nn.Sequential(
depthwise_conv(inp, inp, 3, stride, relu=True), # 3x3可分离卷积
nn.Conv2d(inp, oup, 1, 1, 0, bias=False),
nn.BatchNorm2d(oup),
)
def forward(self, x):
return self.conv(x) + self.shortcut(x) # 相加
class GhostModule(nn.Module):
# ghost模块
def __init__(self, inp, oup, kernel_size=1, ratio=2, dw_size=3, stride=1, relu=True):
super(GhostModule, self).__init__()
self.oup = oup # 输出通道数
init_channels = math.ceil(oup / ratio)
new_channels = init_channels * (ratio - 1) # 计算新的通道数
self.primary_conv = nn.Sequential(
nn.Conv2d(inp, init_channels, kernel_size, stride, kernel_size // 2, bias=False), # 1x1卷积
nn.BatchNorm2d(init_channels),
nn.ReLU(inplace=True) if relu else nn.Sequential(), # relu选择
)
self.cheap_operation = nn.Sequential(
nn.Conv2d(init_channels, new_channels, dw_size, 1, dw_size // 2, groups=init_channels, bias=False), # 3x3分离卷积
nn.BatchNorm2d(new_channels),
nn.ReLU(inplace=True) if relu else nn.Sequential(),
)
def forward(self, x):
x1 = self.primary_conv(x) # 初始卷积计算
x2 = self.cheap_operation(x1) # cheap可分离卷积计算
out = torch.cat([x1, x2], dim=1) # 按通道合并
return out[:, :self.oup, :, :] # 输出前self.oup通道数
def depthwise_conv(inp, oup, kernel_size=3, stride=1, relu=False):
# 可分离卷积
return nn.Sequential(
nn.Conv2d(inp, oup, kernel_size, stride, kernel_size // 2, groups=inp, bias=False),
nn.BatchNorm2d(oup),
nn.ReLU(inplace=True) if relu else nn.Sequential(),
)
class SELayer(nn.Module):
# se模块
def __init__(self, channel, reduction=4):
super(SELayer, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.fc = nn.Sequential(
nn.Linear(channel, channel // reduction),
nn.ReLU(inplace=True),
nn.Linear(channel // reduction, channel), )
def forward(self, x):
b, c, _, _ = x.size()
y = self.avg_pool(x).view(b, c)
y = self.fc(y).view(b, c, 1, 1)
y = torch.clamp(y, 0, 1)
return x * y
论文链接
代码链接