Class ResNet
java.lang.Object
ai.beestream.jitnn.nn.models.ResNet
ResNet (He et al., 2016).
resnet18(int, int, boolean, long) and resnet50(int, int, boolean, long) are torchvision's layouts; the small
variant replaces the 7x7/2 stem and max pool with a 3x3/1 stem for 28-32 pixel inputs (the usual
CIFAR adaptation). Convolutions have no bias, every one is followed by BatchNorm.-
Method Summary
Modifier and TypeMethodDescriptionstatic Sequentialresnet18(int inChannels, int classes, boolean small, long seed) ResNet-18: BasicBlocks [2, 2, 2, 2], widths 64-512.static Sequentialresnet50(int inChannels, int classes, boolean small, long seed) ResNet-50: Bottleneck blocks [3, 4, 6, 3], widths 64-512, expansion 4 (torchvision's v1.5, stride on the 3x3).
-
Method Details
-
resnet18
ResNet-18: BasicBlocks [2, 2, 2, 2], widths 64-512. -
resnet50
ResNet-50: Bottleneck blocks [3, 4, 6, 3], widths 64-512, expansion 4 (torchvision's v1.5, stride on the 3x3).
-