Object Detection on Custom Dataset (Solar Pannels Dataset)
conda env create environment.yml
conda activate obj_detection
python train.py
python test.py
Input as
- Images
- bboxes are in format (Xmin, Ymin, Xmax, Ymax) and unnormalized e.g (420, 171) - (535, 486) not in YOLO format
Using zip for for loop
Using if name == "main"
Building Pytorch project
Using pytorch lighting for training and testing (rewrite the train loop as well as eval loop)
Using wandb
add init.py file