The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more
Add encoder_pool option to gemma4 classification model to toggle soft tokens from encoder vs full sequence
R
Ross Wightman committed
6ce166bd9fe0f000e920ebb864f2d3097493d89d
Parent: eb11943
Committed by Ross Wightman <rwightman@users.noreply.github.com>
on 4/23/2026, 4:57:08 PM