Source code for linear_operator_learning.nn.modules.resnet

"""Resnet Module."""

from typing import Any, Callable, List, Optional, Type, Union

import torch
import torch.nn as nn
from torch import Tensor

__all__ = ["ResNet", "resnet18", "resnet34", "resnet50", "resnet101", "resnet152"]


def conv3x3(
    in_planes: int,
    out_planes: int,
    stride: int = 1,
    groups: int = 1,
    dilation: int = 1,
    padding_mode: str = "zeros",
) -> nn.Conv2d:
    """3x3 convolution with padding."""
    return nn.Conv2d(
        in_planes,
        out_planes,
        kernel_size=3,
        stride=stride,
        padding=dilation,
        groups=groups,
        bias=False,
        dilation=dilation,
        padding_mode=padding_mode,
    )


def conv1x1(in_planes: int, out_planes: int, stride: int = 1) -> nn.Conv2d:
    """1x1 convolution."""
    return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)


class BasicBlock(nn.Module):
    expansion: int = 1

    def __init__(
        self,
        inplanes: int,
        planes: int,
        stride: int = 1,
        downsample: Optional[nn.Module] = None,
        groups: int = 1,
        base_width: int = 64,
        dilation: int = 1,
        padding_mode: str = "zeros",
        norm_layer: Optional[Callable[..., nn.Module]] = None,
    ) -> None:
        super().__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        if groups != 1 or base_width != 64:
            raise ValueError("BasicBlock only supports groups=1 and base_width=64")
        if dilation > 1:
            raise NotImplementedError("Dilation > 1 not supported in BasicBlock")
        # Both self.conv1 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv3x3(inplanes, planes, stride, padding_mode=padding_mode)
        self.bn1 = norm_layer(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes, padding_mode=padding_mode)
        self.bn2 = norm_layer(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x: Tensor) -> Tensor:
        identity = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out


class Bottleneck(nn.Module):
    # Bottleneck in torchvision places the stride for downsampling at 3x3 convolution(self.conv2)
    # while original implementation places the stride at the first 1x1 convolution(self.conv1)
    # according to "Deep residual learning for image recognition" https://arxiv.org/abs/1512.03385.
    # This variant is also known as ResNet V1.5 and improves accuracy according to
    # https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch.

    expansion: int = 4

    def __init__(
        self,
        inplanes: int,
        planes: int,
        stride: int = 1,
        downsample: Optional[nn.Module] = None,
        groups: int = 1,
        base_width: int = 64,
        dilation: int = 1,
        padding_mode: str = "zeros",
        norm_layer: Optional[Callable[..., nn.Module]] = None,
    ) -> None:
        super().__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        width = int(planes * (base_width / 64.0)) * groups
        # Both self.conv2 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv1x1(inplanes, width)
        self.bn1 = norm_layer(width)
        self.conv2 = conv3x3(width, width, stride, groups, dilation, padding_mode=padding_mode)
        self.bn2 = norm_layer(width)
        self.conv3 = conv1x1(width, planes * self.expansion)
        self.bn3 = norm_layer(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x: Tensor) -> Tensor:
        identity = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)
        out = self.relu(out)

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out


[docs] class ResNet(nn.Module): """ResNet model from :footcite:t:`he2016deep`. Args: block (Type[Union[BasicBlock, Bottleneck]]): Block type. layers (List[int]): Number of layers. channels_in (int): Number of input channels. num_features (int): Number of features. zero_init_residual (bool): Zero initialization of residual. groups (int): Number of groups. width_per_group (int): Width per group. replace_stride_with_dilation (Optional[List[bool]]): Replace stride with dilation. padding_mode (str): Padding mode for the convolutional layers. norm_layer (Optional[Callable[..., nn.Module]]): Normalization layer. """ def __init__( self, block: Type[Union[BasicBlock, Bottleneck]], layers: List[int], channels_in: int = 3, num_features: int = 1024, zero_init_residual: bool = False, groups: int = 1, width_per_group: int = 64, replace_stride_with_dilation: Optional[List[bool]] = None, padding_mode: str = "zeros", norm_layer: Optional[Callable[..., nn.Module]] = None, ) -> None: super().__init__() if norm_layer is None: norm_layer = nn.BatchNorm2d self._norm_layer = norm_layer self.inplanes = 64 self.dilation = 1 if replace_stride_with_dilation is None: # each element in the tuple indicates if we should replace # the 2x2 stride with a dilated convolution instead replace_stride_with_dilation = [False, False, False] if len(replace_stride_with_dilation) != 3: raise ValueError( "replace_stride_with_dilation should be None " f"or a 3-element tuple, got {replace_stride_with_dilation}" ) self.groups = groups self.base_width = width_per_group self.conv1 = nn.Conv2d( channels_in, self.inplanes, kernel_size=7, stride=2, padding=3, bias=False, padding_mode=padding_mode, ) self.bn1 = norm_layer(self.inplanes) self.relu = nn.ReLU(inplace=True) self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) self.layer1 = self._make_layer(block, 64, layers[0], padding_mode=padding_mode) self.layer2 = self._make_layer( block, 128, layers[1], stride=2, dilate=replace_stride_with_dilation[0], padding_mode=padding_mode, ) self.layer3 = self._make_layer( block, 256, layers[2], stride=2, dilate=replace_stride_with_dilation[1], padding_mode=padding_mode, ) self.layer4 = self._make_layer( block, 512, layers[3], stride=2, dilate=replace_stride_with_dilation[2], padding_mode=padding_mode, ) self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) self.fc = nn.Linear(512 * block.expansion, num_features, bias=False) 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, nn.GroupNorm)): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) # Zero-initialize the last BN in each residual branch, # so that the residual branch starts with zeros, and each residual block behaves like an identity. # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677 if zero_init_residual: for m in self.modules(): if isinstance(m, Bottleneck) and m.bn3.weight is not None: nn.init.constant_(m.bn3.weight, 0) # type: ignore[arg-type] elif isinstance(m, BasicBlock) and m.bn2.weight is not None: nn.init.constant_(m.bn2.weight, 0) # type: ignore[arg-type] def _make_layer( self, block: Type[Union[BasicBlock, Bottleneck]], planes: int, blocks: int, stride: int = 1, dilate: bool = False, padding_mode: str = "zeros", ) -> nn.Sequential: norm_layer = self._norm_layer downsample = None previous_dilation = self.dilation if dilate: self.dilation *= stride stride = 1 if stride != 1 or self.inplanes != planes * block.expansion: downsample = nn.Sequential( conv1x1(self.inplanes, planes * block.expansion, stride), norm_layer(planes * block.expansion), ) layers = [] layers.append( block( self.inplanes, planes, stride, downsample, self.groups, self.base_width, previous_dilation, padding_mode, norm_layer, ) ) self.inplanes = planes * block.expansion for _ in range(1, blocks): layers.append( block( self.inplanes, planes, groups=self.groups, base_width=self.base_width, dilation=self.dilation, norm_layer=norm_layer, padding_mode=padding_mode, ) ) return nn.Sequential(*layers) def _forward_impl(self, x: Tensor) -> Tensor: # See note [TorchScript super()] x = self.conv1(x) x = self.bn1(x) x = self.relu(x) x = self.maxpool(x) x = self.layer1(x) x = self.layer2(x) x = self.layer3(x) x = self.layer4(x) x = self.avgpool(x) x = torch.flatten(x, 1) x = self.fc(x) return x def forward(self, x: Tensor) -> Tensor: """Forward pass of the ResNet model.""" return self._forward_impl(x)
def _resnet( block: Type[Union[BasicBlock, Bottleneck]], layers: List[int], **kwargs: Any, ) -> ResNet: model = ResNet(block, layers, **kwargs) return model def resnet18(*args, **kwargs: Any) -> ResNet: """ResNet-18 from `Deep Residual Learning for Image Recognition <https://arxiv.org/abs/1512.03385>`__.""" return _resnet(BasicBlock, [2, 2, 2, 2], **kwargs) def resnet34(*args, **kwargs: Any) -> ResNet: """ResNet-34 from `Deep Residual Learning for Image Recognition <https://arxiv.org/abs/1512.03385>`__.""" return _resnet(BasicBlock, [3, 4, 6, 3], **kwargs) def resnet50(*args, **kwargs: Any) -> ResNet: """ResNet-50 from `Deep Residual Learning for Image Recognition <https://arxiv.org/abs/1512.03385>`__.""" return _resnet(Bottleneck, [3, 4, 6, 3], **kwargs) def resnet101(*args, **kwargs: Any) -> ResNet: """ResNet-101 from `Deep Residual Learning for Image Recognition <https://arxiv.org/abs/1512.03385>`__.""" return _resnet(Bottleneck, [3, 4, 23, 3], **kwargs) def resnet152(*args, **kwargs: Any) -> ResNet: """ResNet-101 from `Deep Residual Learning for Image Recognition <https://arxiv.org/abs/1512.03385>`__.""" return _resnet(Bottleneck, [3, 8, 36, 3], **kwargs)