Skip to content

Latest commit

 

History

History
25 lines (18 loc) · 968 Bytes

README.md

File metadata and controls

25 lines (18 loc) · 968 Bytes
Raw

Transferring to Detection

The train_object_detection.py script reproduces the object detection experiments on Pascal VOC and COCO.

Instruction

  1. Install detectron2.

  2. Convert a pre-trained model to detectron2's format:

    python3 convert_model_to_detectron2.py --pretrained_feature_extractor PATH_TO_CKPT --output_detectron_model ./detectron_model.pkl

  3. Put dataset under "./datasets" directory, following the directory structure requried by detectron2.

  4. Run training:

    python train_net.py --config-file configs/pascal_voc_R_50_C4_24k_moco.yaml \
 --num-gpus 8 MODEL.WEIGHTS ./detectron_model.pkl