A fork of detectron2 contains the code implementation of detection and instance segmentation experiments in ResNeSt.

We train all models with FPN, SyncBN and image scale augmentation (short size of a image is pickedrandomly from 640 to 800). 1x learning rate schedule is used. All of them are reported on COCO-2017 validation dataset.

All models are trained along with FPN and SyncBN. For data augmentation,input images’ shorter side are randomly scaled to one of (640, 672, 704, 736, 768, 800). 1x learning rate schedule is used, if not otherwise specified. All of them are reported on COCO-2017 validation dataset. The values with * demonstrate the mutli-scale testing performance on the test-dev2019.

Please follow INSTALL.md to install detecron2.

To train a model with 8 gpus, please run

python tools/train_net.py  --num-gpus 8 --config-file your_config.yaml

For inference

python tools/train_net.py  \
                --config-file your_config.yaml
                --eval-only MODEL.WEIGHTS /path/to/checkpoint_file

For the inference demo, please see GETTING_STARTED.md.

ResNeSt: Split-Attention Networks [arXiv]

Hang Zhang, Chongruo Wu, Zhongyue Zhang, Yi Zhu, Zhi Zhang, Haibin Lin, Yue Sun, Tong He, Jonas Muller, R. Manmatha, Mu Li and Alex Smola

@article{zhang2020resnest,
title={ResNeSt: Split-Attention Networks},
author={Zhang, Hang and Wu, Chongruo and Zhang, Zhongyue and Zhu, Yi and Zhang, Zhi and Lin, Haibin and Sun, Yue and He, Tong and Muller, Jonas and Manmatha, R. and Li, Mu and Smola, Alexander},
journal={arXiv preprint arXiv:2004.08955},
year={2020}
}

• ResNeSt (main)
• ResNeSt on ImageNet (Gluon, PyTorch)
• ResNeSt on Semantic Segmentation (Gluon, PyTorch)

Chongruo Wu, Zhongyue Zhang, Hang Zhang