Quick build detectron on slurm

• bash rebuild_detectron2.sh

Download weights

• R-FPN ResNet50 trained with C-RBP for 20 steps: wget https://bashupload.com/qyTMQ/-D4Ro.pth
• R-FPN ResNet101 trained with C-RBP for 20 steps: wget https://bashupload.com/V_Rhr/lwUiz.pth
• FPN ResNet50: wget https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_50_1x/139514544/model_final_dbfeb4.pkl
• FPN ResNet101: wget https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_101_3x/139514519/model_final_cafdb1.pkl

Train a model

• R-FPN ResNet50 trained with C-RBP for 20 steps: python tools/train_net.py --num-gpus 10 --config-file configs/COCO-PanopticSegmentation/panoptic_rfpn_R_50_cbp20_1x.yaml SOLVER.IMS_PER_BATCH 40 SOLVER.BASE_LR 0.05

Test a model

• R-FPN ResNet50 trained with C-RBP for 20 steps: python tools/train_net.py --eval-only --num-gpus 10 --config-file configs/COCO-PanopticSegmentation/panoptic_rfpn_R_50_cbp20_1x-test.yaml MODEL.WEIGHTS <path-to-weights>

---

Detectron2 is Facebook AI Research's next generation software system that implements state-of-the-art object detection algorithms. It is a ground-up rewrite of the previous version, Detectron, and it originates from maskrcnn-benchmark.

What's New

• It is powered by the PyTorch deep learning framework.
• Includes more features such as panoptic segmentation, densepose, Cascade R-CNN, rotated bounding boxes, etc.
• Can be used as a library to support different projects on top of it. We'll open source more research projects in this way.
• It trains much faster.

See our blog post to see more demos and learn about detectron2.

Installation

See INSTALL.md.

Quick Start

See GETTING_STARTED.md, or the Colab Notebook.

Learn more at our documentation. And see projects/ for some projects that are built on top of detectron2.

Model Zoo and Baselines

We provide a large set of baseline results and trained models available for download in the Detectron2 Model Zoo.

License

Detectron2 is released under the Apache 2.0 license.

Citing Detectron

If you use Detectron2 in your research or wish to refer to the baseline results published in the Model Zoo, please use the following BibTeX entry.

@misc{wu2019detectron2,
  author =       {Yuxin Wu and Alexander Kirillov and Francisco Massa and
                  Wan-Yen Lo and Ross Girshick},
  title =        {Detectron2},
  howpublished = {\url{https://github.com/facebookresearch/detectron2}},
  year =         {2019}
}