A CV toolkit for my papers.
MIT License
Bot releases are hidden (Show)
Change model zoo to wasabi to save model
Published by zhanghang1989 over 4 years ago
Split-Attention Network, A New ResNet Variant. It significantly boosts the performance of downstream models such as Mask R-CNN, Cascade R-CNN and DeepLabV3.
crop size | PyTorch | Gluon | |
---|---|---|---|
ResNeSt-50 | 224 | 81.03 | 81.04 |
ResNeSt-101 | 256 | 82.83 | 82.81 |
ResNeSt-200 | 320 | 83.84 | 83.88 |
ResNeSt-269 | 416 | 84.54 | 84.53 |
Published by zhanghang1989 over 4 years ago
Published by zhanghang1989 almost 6 years ago
Published by zhanghang1989 over 6 years ago
pip install torch-encoding
Published by zhanghang1989 almost 7 years ago
Make the compilation easier to the users.
Published by zhanghang1989 about 7 years ago
Memory efficient implementation of Encoding Layer.