Referring Tweets

@alxndrkalinin #PyTorch implementation and pretrained models for RepVGG: Making VGG-style ConvNets Great Again - multi-branch topology at training - simple 3x3 convs & ReLU at inference - >80% top-1 accuracy on ImageNet - 83% faster than ResNet-50 on NVIDIA 1080Ti
@ak92501 RepVGG: Making VGG-style ConvNets Great Again pdf: abs: github:
@s1ddok I enjoy reading papers that focus on moving hard/slow things to offline/training stage. This simple re-parameterization technique is so awesome, It would be fun to use this in StyleGAN2 given that it basically features almost the same plain-stack design
@ZFPhalanx RepVGG: Making VGG-style ConvNets Great Again

Bookmark Comments

Related Entries

Read more GitHub - MishaLaskin/rad: RAD: Reinforcement Learning with Augmented Data
0 users, 8 mentions 2020/05/01 02:21
Read more GitHub - facebookresearch/consistent_depth: We estimate dense, flicker-free, geometrically consisten...
0 users, 15 mentions 2020/08/18 08:21
Read more GitHub - microsoft/Bringing-Old-Photos-Back-to-Life: Bringing Old Photo Back to Life (CVPR 2020 oral...
0 users, 13 mentions 2020/09/17 17:21
Read more GitHub - chaofengc/PSFRGAN: PyTorch codes for "Progressive Semantic-Aware Style Transformation for B...
0 users, 4 mentions 2020/09/24 14:21
Read more GitHub - alexa/bort: Repository for the paper "Optimal Subarchitecture Extraction for BERT"
0 users, 9 mentions 2020/10/21 14:21