"""
Modified from https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv2.py
"""
import math
from typing import Callable, Any, Optional, List
import oneflow as flow
from oneflow import nn, Tensor
from .utils import load_state_dict_from_url
from .registry import ModelCreator
from .helpers import make_divisible
__all__ = ["MobileNetV2", "mobilenet_v2"]
model_urls = {
"mobilenet_v2": "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/MobileNet/mobilenet_v2.zip",
}
class ConvBNActivation(nn.Sequential):
def __init__(
self,
in_planes: int,
out_planes: int,
kernel_size: int = 3,
stride: int = 1,
groups: int = 1,
norm_layer: Optional[Callable[..., nn.Module]] = None,
activation_layer: Optional[Callable[..., nn.Module]] = None,
dilation: int = 1,
) -> None:
padding = (kernel_size - 1) // 2 * dilation
if norm_layer is None:
norm_layer = nn.BatchNorm2d
if activation_layer is None:
activation_layer = nn.ReLU6
super().__init__(
nn.Conv2d(
in_planes,
out_planes,
kernel_size,
stride,
padding,
dilation=dilation,
groups=groups,
bias=False,
),
norm_layer(out_planes),
activation_layer(inplace=True),
)
self.out_channels = out_planes
# necessary for backwards compatibility
ConvBNReLU = ConvBNActivation
class InvertedResidual(nn.Module):
def __init__(
self,
inp: int,
oup: int,
stride: int,
expand_ratio: int,
norm_layer: Optional[Callable[..., nn.Module]] = None,
) -> None:
super(InvertedResidual, self).__init__()
self.stride = stride
assert stride in [1, 2]
if norm_layer is None:
norm_layer = nn.BatchNorm2d
hidden_dim = int(round(inp * expand_ratio))
self.use_res_connect = self.stride == 1 and inp == oup
layers: List[nn.Module] = []
if expand_ratio != 1:
# pw
layers.append(
ConvBNReLU(inp, hidden_dim, kernel_size=1, norm_layer=norm_layer)
)
layers.extend(
[
# dw
ConvBNReLU(
hidden_dim,
hidden_dim,
stride=stride,
groups=hidden_dim,
norm_layer=norm_layer,
),
# pw-linear
nn.Conv2d(hidden_dim, oup, 1, 1, 0, bias=False),
norm_layer(oup),
]
)
self.conv = nn.Sequential(*layers)
self.out_channels = oup
self._is_cn = stride > 1
def forward(self, x: flow.Tensor) -> flow.Tensor:
if self.use_res_connect:
return x + self.conv(x)
else:
return self.conv(x)
class MobileNetV2(nn.Module):
def __init__(
self,
num_classes: int = 1000,
width_mult: float = 1.0,
inverted_residual_setting: Optional[List[List[int]]] = None,
round_nearest: int = 8,
block: Optional[Callable[..., nn.Module]] = None,
norm_layer: Optional[Callable[..., nn.Module]] = None,
) -> None:
"""
MobileNet V2 main class
Args:
num_classes (int): Number of classes
width_mult (float): Width multiplier - adjusts number of channels in each layer by this amount
inverted_residual_setting: Network structure
round_nearest (int): Round the number of channels in each layer to be a multiple of this number.
Set to 1 to turn off rounding
block: Module specifying inverted residual building block for mobilenet
norm_layer: Module specifying the normalization layer to use
"""
super(MobileNetV2, self).__init__()
if block is None:
block = InvertedResidual
if norm_layer is None:
norm_layer = nn.BatchNorm2d
input_channel = 32
last_channel = 1280
if inverted_residual_setting is None:
inverted_residual_setting = [
# t, c, n, s
[1, 16, 1, 1],
[6, 24, 2, 2],
[6, 32, 3, 2],
[6, 64, 4, 2],
[6, 96, 3, 1],
[6, 160, 3, 2],
[6, 320, 1, 1],
]
# only check the first element, assuming user knows t,c,n,s are required
if (
len(inverted_residual_setting) == 0
or len(inverted_residual_setting[0]) != 4
):
raise ValueError(
"inverted_residual_setting should be non-empty "
"or a 4-element list, got {}".format(inverted_residual_setting)
)
# building first layer
input_channel = make_divisible(input_channel * width_mult, round_nearest)
self.last_channel = make_divisible(
last_channel * max(1.0, width_mult), round_nearest
)
features: List[nn.Module] = [
ConvBNReLU(3, input_channel, stride=2, norm_layer=norm_layer)
]
# building inverted residual blocks
for t, c, n, s in inverted_residual_setting:
output_channel = make_divisible(c * width_mult, round_nearest)
for i in range(n):
stride = s if i == 0 else 1
features.append(
block(
input_channel,
output_channel,
stride,
expand_ratio=t,
norm_layer=norm_layer,
)
)
input_channel = output_channel
# building last several layers
features.append(
ConvBNReLU(
input_channel, self.last_channel, kernel_size=1, norm_layer=norm_layer
)
)
# make it nn.Sequential
self.features = nn.Sequential(*features)
self.adaptive_avg_pool2d = nn.AdaptiveAvgPool2d((1, 1))
# building classifier
self.classifier = nn.Sequential(
nn.Dropout(0.2), nn.Linear(self.last_channel, num_classes),
)
# weight initialization
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode="fan_out")
if m.bias is not None:
nn.init.zeros_(m.bias)
elif isinstance(m, (nn.BatchNorm2d)):
nn.init.ones_(m.weight)
nn.init.zeros_(m.bias)
elif isinstance(m, nn.Linear):
nn.init.normal_(m.weight, 0, 0.01)
nn.init.zeros_(m.bias)
def _forward_impl(self, x: Tensor) -> Tensor:
x = self.features(x)
# Cannot use "squeeze" as batch-size can be 1
x = self.adaptive_avg_pool2d(x)
x = flow.flatten(x, 1)
x = self.classifier(x)
return x
def forward(self, x: Tensor) -> Tensor:
return self._forward_impl(x)
[docs]@ModelCreator.register_model
def mobilenet_v2(
pretrained: bool = False, progress: bool = True, **kwargs: Any
) -> MobileNetV2:
"""
Constructs the MobileNetV2 model.
.. note::
MobileNetV2 model architecture from the `MobileNetV2: Inverted Residuals and Linear Bottlenecks <https://arxiv.org/abs/1801.04381>`_ paper.
Args:
pretrained (bool): Whether to download the pre-trained model on ImageNet. Default: ``False``
progress (bool): If True, displays a progress bar of the download to stderr. Default: ``True``
For example:
.. code-block:: python
>>> import flowvision
>>> mobilenet_v2 = flowvision.models.mobilenet_v2(pretrained=False, progress=True)
"""
model = MobileNetV2(**kwargs)
if pretrained:
state_dict = load_state_dict_from_url(
model_urls["mobilenet_v2"], progress=progress
)
model.load_state_dict(state_dict)
return model