Implement dynamic num_classes and resolve dimension mismatch in ResNet18 #91

ResNet.py

@@ -0,0 +1,66 @@
+import torch
+from torch import nn
+
+class BasicBlock(nn.Module):
+    expansion = 1
+    def __init__(self, in_channels, out_channels, stride=1):
+        super(BasicBlock, self).__init__()
+        self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1, bias=False)
+        self.bn1 = nn.BatchNorm2d(out_channels)
+        self.relu = nn.ReLU(inplace=True)
+        self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1, bias=False)
+        self.bn2 = nn.BatchNorm2d(out_channels)
+        self.stride = stride
+        self.shortcut_conv = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=stride, bias=False)
+        self.shortcut_bn = nn.BatchNorm2d(out_channels)
+
+    def forward(self, x):
+        identity = x
+        out = self.conv1(x)
+        out = self.bn1(out)
+        out = self.relu(out)
+        out = self.conv2(out)
+        out = self.bn2(out)
+        if self.stride != 1 or identity.size() != out.size():
+            identity = self.shortcut_conv(identity)
+            identity = self.shortcut_bn(identity)
+        out += identity
+        out = self.relu(out)
+        return out
+
+class ResNet18(nn.Module):
+    def __init__(self, num_classes=1000):
+        super(ResNet18, self).__init__()
+        self.in_channels = 64
+        self.conv1 = nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3, bias=False)
+        self.bn1 = nn.BatchNorm2d(64)
+        self.relu = nn.ReLU(inplace=True)
+        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
+        self.layer1 = self._make_layer(BasicBlock, 64, 2, stride=1)
+        self.layer2 = self._make_layer(BasicBlock, 128, 2, stride=2)
+        self.layer3 = self._make_layer(BasicBlock, 256, 2, stride=2)
+        self.layer4 = self._make_layer(BasicBlock, 512, 2, stride=1)
+        self.avgpool = nn.AdaptiveAvgPool2d((1, 32))
+        self.fc = nn.Linear(512 * 32, num_classes)
+
+    def _make_layer(self, block, out_channels, num_blocks, stride):
+        strides = [stride] + [1] * (num_blocks - 1)
+        layers = nn.ModuleList()
+        for stride in strides:
+            layers.append(block(self.in_channels, out_channels, stride))
+            self.in_channels = out_channels * block.expansion
+        return layers
+
+    def forward(self, x):
+        x = self.conv1(x)
+        x = self.bn1(x)
+        x = self.relu(x)
+        x = self.maxpool(x)
+        for layer in self.layer1: x = layer(x)
+        for layer in self.layer2: x = layer(x)
+        for layer in self.layer3: x = layer(x)
+        for layer in self.layer4: x = layer(x)
+        x = self.avgpool(x)
+        x = torch.flatten(x, 1)
+        x = self.fc(x)
+        return x