import torch
import torch.nn as nn


class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, planes, stride=1):
        super(BasicBlock, self).__init__()
        self.conv1 = nn.Conv2d(
            in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False
        )
        self.bn1 = nn.BatchNorm2d(planes)
        self.conv2 = nn.Conv2d(
            planes, planes, kernel_size=3, stride=1, padding=1, bias=False
        )
        self.bn2 = nn.BatchNorm2d(planes)

        self.shortcut = nn.Sequential()
        if stride != 1 or in_planes != self.expansion * planes:
            self.shortcut = nn.Sequential(
                nn.Conv2d(
                    in_planes,
                    self.expansion * planes,
                    kernel_size=1,
                    stride=stride,
                    bias=False,
                ),
                nn.BatchNorm2d(self.expansion * planes),
            )

    def forward(self, x):
        out = torch.relu(self.bn1(self.conv1(x)))
        out = self.bn2(self.conv2(out))
        out += self.shortcut(x)
        out = torch.relu(out)
        return out


class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, in_planes, planes, stride=1):
        super(Bottleneck, self).__init__()
        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=1, bias=False)
        self.bn1 = nn.BatchNorm2d(planes)
        self.conv2 = nn.Conv2d(
            planes, planes, kernel_size=3, stride=stride, padding=1, bias=False
        )
        self.bn2 = nn.BatchNorm2d(planes)
        self.conv3 = nn.Conv2d(
            planes, self.expansion * planes, kernel_size=1, bias=False
        )
        self.bn3 = nn.BatchNorm2d(self.expansion * planes)

        self.shortcut = nn.Sequential()
        if stride != 1 or in_planes != self.expansion * planes:
            self.shortcut = nn.Sequential(
                nn.Conv2d(
                    in_planes,
                    self.expansion * planes,
                    kernel_size=1,
                    stride=stride,
                    bias=False,
                ),
                nn.BatchNorm2d(self.expansion * planes),
            )

    def forward(self, x):
        out = torch.relu(self.bn1(self.conv1(x)))
        out = torch.relu(self.bn2(self.conv2(out)))
        out = self.bn3(self.conv3(out))
        out += self.shortcut(x)
        out = torch.relu(out)
        return out


class ResNet(nn.Module):
    def __init__(self, block, num_blocks, num_classes=1000, K=10, T=0.5):
        super(ResNet, self).__init__()
        self.in_planes = 64
        self.K = K
        self.T = T

        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
        self.bn1 = nn.BatchNorm2d(64)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
        self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
        self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2)
        self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2)
        self.fc = nn.Linear(512 * block.expansion, num_classes)

    def _make_layer(self, block, planes, num_blocks, stride):
        strides = [stride] + [1] * (num_blocks - 1)
        layers = []
        for stride in strides:
            layers.append(block(self.in_planes, planes, stride))
            self.in_planes = planes * block.expansion
        return nn.Sequential(*layers)

    def t_max_avg_pooling(self, x):
        B, C, H, W = x.shape
        x_flat = x.view(B, C, -1)
        top_k_values, _ = torch.topk(x_flat, self.K, dim=2)
        max_values = top_k_values.max(dim=2)[0]
        avg_values = top_k_values.mean(dim=2)
        output = torch.where(max_values >= self.T, max_values, avg_values)
        return output

    def forward(self, x):
        out = torch.relu(self.bn1(self.conv1(x)))
        out = self.maxpool(out)
        out = self.layer1(out)
        out = self.layer2(out)
        out = self.layer3(out)
        out = self.layer4(out)
        out = self.t_max_avg_pooling(out)
        out = out.view(out.size(0), -1)
        out = self.fc(out)
        return out


def ResNet18(num_classes=1000, K=10, T=0.5):
    return ResNet(BasicBlock, [2, 2, 2, 2], num_classes, K, T)


def ResNet34(num_classes=1000, K=10, T=0.5):
    return ResNet(BasicBlock, [3, 4, 6, 3], num_classes, K, T)


def ResNet50(num_classes=1000, K=10, T=0.5):
    return ResNet(Bottleneck, [3, 4, 6, 3], num_classes, K, T)


def ResNet101(num_classes=1000, K=10, T=0.5):
    return ResNet(Bottleneck, [3, 4, 23, 3], num_classes, K, T)


def ResNet152(num_classes=1000, K=10, T=0.5):
    return ResNet(Bottleneck, [3, 8, 36, 3], num_classes, K, T)