exp/baseline3/model.py

import torch
import torch.nn as nn
from huggingface_hub import PyTorchModelHubMixin


class SEBlock(nn.Module):
    def __init__(self, channels, reduction=16):
        super().__init__()
        self.avg_pool = nn.AdaptiveAvgPool2d(1)
        self.fc = nn.Sequential(
            nn.Linear(channels, channels // reduction, bias=False),
            nn.ReLU(inplace=True),
            nn.Linear(channels // reduction, channels, bias=False),
            nn.Sigmoid(),
        )

    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)
        return x * y.expand_as(x)


class SEResNeXtBlock(nn.Module):
    expansion = 2

    def __init__(self, in_channels, planes, stride=1, downsample=None, groups=8, base_width=4):
        super().__init__()
        # Calculate width based on ResNeXt formula
        # width = floor(planes * (base_width/64)) * groups
        # For small planes, this might be 0. Let's ensure minimum width.
        width = int(planes * (base_width / 64.0)) * groups
        if width < groups:
            width = groups
        
        self.conv1 = nn.Conv2d(in_channels, width, kernel_size=1, bias=False)
        self.bn1 = nn.BatchNorm2d(width)
        
        self.conv2 = nn.Conv2d(width, width, kernel_size=3, stride=stride, padding=1, groups=groups, bias=False)
        self.bn2 = nn.BatchNorm2d(width)
        
        self.conv3 = nn.Conv2d(width, planes * self.expansion, kernel_size=1, bias=False)
        self.bn3 = nn.BatchNorm2d(planes * self.expansion)
        
        self.relu = nn.ReLU(inplace=True)
        self.se = SEBlock(planes * self.expansion)
        self.downsample = downsample

    def forward(self, x):
        identity = x
        if self.downsample is not None:
            identity = self.downsample(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)
        out = self.se(out)

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


class SEResNeXt(nn.Module, PyTorchModelHubMixin):
    def __init__(
        self, layers=[2, 2, 2, 2], planes=[16, 32, 64, 128], dropout_rate=0.5, groups=8, base_width=4
    ):
        super().__init__()
        self.in_channels = 32  # Increased stem size
        self.groups = groups
        self.base_width = base_width

        # Stem
        self.conv1 = nn.Conv2d(3, 32, kernel_size=3, stride=1, padding=1, bias=False)
        self.bn1 = nn.BatchNorm2d(32)
        self.relu = nn.ReLU(inplace=True)

        # Stages
        self.layer1 = self._make_layer(planes[0], layers[0], stride=1)
        self.layer2 = self._make_layer(planes[1], layers[1], stride=2)
        self.layer3 = self._make_layer(planes[2], layers[2], stride=2)
        self.layer4 = self._make_layer(planes[3], layers[3], stride=2)

        self.dropout = nn.Dropout(p=dropout_rate)

        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        # Final channel count is planes[3] * expansion (2)
        self.fc = nn.Linear(planes[3] * SEResNeXtBlock.expansion, 1)
        self.sigmoid = nn.Sigmoid()

    def _make_layer(self, planes, blocks, stride=1):
        downsample = None
        out_channels = planes * SEResNeXtBlock.expansion
        
        if stride != 1 or self.in_channels != out_channels:
            downsample = nn.Sequential(
                nn.Conv2d(
                    self.in_channels,
                    out_channels,
                    kernel_size=1,
                    stride=stride,
                    bias=False,
                ),
                nn.BatchNorm2d(out_channels),
            )

        layers = []
        layers.append(SEResNeXtBlock(self.in_channels, planes, stride, downsample, self.groups, self.base_width))
        self.in_channels = out_channels
        for _ in range(1, blocks):
            layers.append(SEResNeXtBlock(self.in_channels, planes, groups=self.groups, base_width=self.base_width))

        return nn.Sequential(*layers)

    def forward(self, x):
        # x: (B, 3, 80, 101)
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(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.dropout(x)
        x = self.fc(x)
        x = self.sigmoid(x)

        return x


if __name__ == "__main__":
    from torchinfo import summary

    model = SEResNeXt()
    summary(model, (1, 3, 80, 101))