| import torch |
| import torch.nn as nn |
| import torch.nn.functional as F |
| from typing import List |
| import logging |
|
|
| from config import ModelConfig |
|
|
| logger = logging.getLogger(__name__) |
|
|
| 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 = F.relu(self.bn1(self.conv1(x))) |
| out = self.bn2(self.conv2(out)) |
| out += self.shortcut(x) |
| out = F.relu(out) |
| return out |
|
|
|
|
| class ResNet(nn.Module): |
| def __init__(self, block, num_blocks: List[int], config: ModelConfig): |
| super(ResNet, self).__init__() |
| self.config = config |
| self.in_planes = 64 |
|
|
| self.conv1 = nn.Conv2d(config.input_channels, 64, kernel_size=3, stride=1, padding=1, bias=False) |
| self.bn1 = nn.BatchNorm2d(64) |
| 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.linear = nn.Linear(512 * block.expansion, config.num_classes) |
| |
| logger.info(f"Created {config.name} with {config.num_classes} 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 forward(self, x): |
| out = F.relu(self.bn1(self.conv1(x))) |
| out = self.layer1(out) |
| out = self.layer2(out) |
| out = self.layer3(out) |
| out = self.layer4(out) |
| out = F.avg_pool2d(out, 4) |
| out = out.view(out.size(0), -1) |
| out = self.linear(out) |
| return out |
|
|
|
|
| def ResNet18(config: ModelConfig = None): |
| """Create ResNet-18 model""" |
| if config is None: |
| config = ModelConfig() |
| return ResNet(BasicBlock, [2, 2, 2, 2], config) |
|
|
|
|
| def ResNet34(config: ModelConfig = None): |
| """Create ResNet-34 model""" |
| if config is None: |
| config = ModelConfig() |
| return ResNet(BasicBlock, [3, 4, 6, 3], config) |
|
|
| def create_model(config: ModelConfig) -> nn.Module: |
| """Factory function to create models based on configuration""" |
| models = { |
| 'ResNet18': ResNet18, |
| 'ResNet34': ResNet34, |
| } |
| |
| if config.name not in models: |
| raise ValueError(f"Unknown model: {config.name}. Available: {list(models.keys())}") |
| |
| return models[config.name](config) |
|
|
