train the STD Model on my own dataset

quxianjiuguo commented 6 months ago

Congradulations for accepted work Is this git complete? I would like to train your suggested Spatial-Transform-Decoupling model on my own dataset. However, I did not find any code related to your suggested Spatial-Transform-Decoupling model. Could you provide a more detailed user guide？ Thanks a lot

yuhongtian17 commented 6 months ago

Our core code: bbox_heads, cfgs.

Indeed, Installation in README only tell readers how to use rotated_faster_rcnn+r50 in MMRotate. Further, you can use our rotated_faster_rcnn+vit_small+std like:

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 ./tools/dist_train.sh ./configs/rotated_imted/rotated_imted_faster_rcnn_vit_small_1x_dota_le90_8h_stdc_xyawh321v.py 8
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 ./tools/dist_test.sh ./configs/rotated_imted/rotated_imted_faster_rcnn_vit_small_1x_dota_le90_8h_stdc_xyawh321v.py ./work_dirs/rotated_imted_faster_rcnn_vit_small_1x_dota_le90_8h_stdc_xyawh321v/epoch_12.pth 8 --format-only --eval-options submission_dir="./work_dirs/Task1_rotated_imted_faster_rcnn_vit_small_1x_dota_le90_8h_stdc_xyawh321v_epoch_12/"
python "../DOTA_devkit/dota_evaluation_task1.py" --mergedir "./work_dirs/Task1_rotated_imted_faster_rcnn_vit_small_1x_dota_le90_8h_stdc_xyawh321v_epoch_12/" --imagesetdir "./data/DOTA/val/" --use_07_metric True

yaosx425 commented 5 months ago

对已接受工作的祝贺这个 git 完成吗？我想在我自己的数据集上训练您建议的空间变换解耦模型。但是，我没有找到与您建议的空间变换解耦模型相关的任何代码。您能提供更详细的用户指南吗？非常感谢

Hello, classmate! Have you run successfully on your own dataset?

yuhongtian17 / Spatial-Transform-Decoupling

train the STD Model on my own dataset #6