aot/networks/encoders/mobilenetv2.py

from torch import nn
from torch import Tensor
from typing import Callable, Optional, List
from utils.learning import freeze_params

__all__ = ['MobileNetV2']


def _make_divisible(v: float,
                    divisor: int,
                    min_value: Optional[int] = None) -> int:
    """
    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 ConvBNActivation(nn.Sequential):
    def __init__(
        self,
        in_planes: int,
        out_planes: int,
        kernel_size: int = 3,
        stride: int = 1,
        groups: int = 1,
        padding: int = -1,
        norm_layer: Optional[Callable[..., nn.Module]] = None,
        activation_layer: Optional[Callable[..., nn.Module]] = None,
        dilation: int = 1,
    ) -> None:
        if padding == -1:
            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,
            dilation: 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

        self.kernel_size = 3
        self.dilation = dilation

        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,
                       dilation=dilation,
                       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: Tensor) -> Tensor:
        if self.use_res_connect:
            return x + self.conv(x)
        else:
            return self.conv(x)


class MobileNetV2(nn.Module):
    def __init__(self,
                 output_stride=8,
                 norm_layer: Optional[Callable[..., nn.Module]] = None,
                 width_mult: float = 1.0,
                 inverted_residual_setting: Optional[List[List[int]]] = None,
                 round_nearest: int = 8,
                 block: Optional[Callable[..., nn.Module]] = None,
                 freeze_at=0) -> 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

        last_channel = 1280
        input_channel = 32
        current_stride = 1
        rate = 1

        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)
        ]
        current_stride *= 2
        # building inverted residual blocks
        for t, c, n, s in inverted_residual_setting:
            if current_stride == output_stride:
                stride = 1
                dilation = rate
                rate *= s
            else:
                stride = s
                dilation = 1
                current_stride *= s
            output_channel = _make_divisible(c * width_mult, round_nearest)
            for i in range(n):
                if i == 0:
                    features.append(
                        block(input_channel, output_channel, stride, dilation,
                              t, norm_layer))
                else:
                    features.append(
                        block(input_channel, output_channel, 1, rate, t,
                              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._initialize_weights()

        feature_4x = self.features[0:4]
        feautre_8x = self.features[4:7]
        feature_16x = self.features[7:14]
        feature_32x = self.features[14:]

        self.stages = [feature_4x, feautre_8x, feature_16x, feature_32x]

        self.freeze(freeze_at)

    def forward(self, x):
        xs = []
        for stage in self.stages:
            x = stage(x)
            xs.append(x)
        return xs

    def _initialize_weights(self):
        # 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.GroupNorm)):
                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 freeze(self, freeze_at):
        if freeze_at >= 1:
            for m in self.stages[0][0]:
                freeze_params(m)

        for idx, stage in enumerate(self.stages, start=2):
            if freeze_at >= idx:
                freeze_params(stage)