CNNs

Convolutional Neural Networks

Computer Vision Backbones Vision Models Convolutional Zoo

Contrastive Pretraining for CNNs

• GitHub - keyu-tian/SparK: [ICLR’23 Spotlight] The first successful BERT-style pretraining on any convolutional network; Pytorch impl. of “Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling”

Fast Training

• GitHub - tysam-code/hlb-CIFAR10: Train to 94% on CIFAR-10 in less than 10 seconds on a single A100, the current world record. Or ~95.77% in ~188 seconds.
• GitHub - KellerJordan/cifar10-airbench: 94% on CIFAR-10 in 2.73 seconds 💨 96% in 27 seconds