ResNetVariant
Enums · AiDotNet.Enums
Defines the available ResNet (Residual Network) architecture variants.
For Beginners
ResNet networks are named after their total number of weight layers. For example, ResNet50 has 50 convolutional and fully-connected layers. These networks can be much deeper than earlier architectures (like VGG) because the skip connections allow gradients to flow more easily during training, solving the “vanishing gradient” problem.
How It Works
ResNet architectures are a family of deep convolutional neural networks that introduced skip connections (residual connections) to enable training of very deep networks. The key innovation is learning residual functions with reference to the layer inputs, rather than learning unreferenced functions.
Architecture Types:
- ResNet18/34 use “BasicBlock” with two 3x3 convolutions
- ResNet50/101/152 use “BottleneckBlock” with 1x1, 3x3, 1x1 convolutions for efficiency
Fields
| Field | Summary |
|---|---|
ResNetVariant ResNet101 |
ResNet-101: 101 weight layers using BottleneckBlock. |
ResNetVariant ResNet152 |
ResNet-152: 152 weight layers using BottleneckBlock. |
ResNetVariant ResNet18 |
ResNet-18: 18 weight layers using BasicBlock. |
ResNetVariant ResNet34 |
ResNet-34: 34 weight layers using BasicBlock. |
ResNetVariant ResNet50 |
ResNet-50: 50 weight layers using BottleneckBlock. |
int value__ |