yjh0410 5bcd4bc7e5 train YOLOv8-S with 300 epoch		1 vuosi sitten
..
basic_modules	5bcd4bc7e5 train YOLOv8-S with 300 epoch	1 vuosi sitten
README.md	9fcfcadcae update	1 vuosi sitten
build.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
loss.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
loss_utils.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
matcher.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
rtdetr.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
rtdetr_decoder.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten
rtdetr_encoder.py	264112178f build a new YOLO-Tutorial project for my book	1 vuosi sitten

Real-time Transformer-based Object Detector:

Results on the COCO-val

Taking training RT-DETR-S on COCO as the example,

python train.py --cuda -d coco --root path/to/coco -m rtdetr_r18 -bs 16  --fp16

Taking training RT-DETR on COCO as the example,

python -m torch.distributed.run --nproc_per_node=8 train.py --cuda --distributed -d coco --root /data/datasets/ -m rtdetr_r18 -bs 16 --fp16 --sybn

Taking testing RT-DETR on COCO-val as the example,

python test.py --cuda -d coco --root path/to/coco -m rtdetr_r18 --weight path/to/rtdetr_r18.pth --show

Taking evaluating RT-DETR on COCO-val as the example,

python train.py --cuda -d coco --root path/to/coco -m rtdetr_r18 -bs 16 --fp16 --resume path/to/rtdetr_r18.pth --eval_first

python demo.py --mode image --path_to_img path/to/image_dirs/ --cuda -m rtdetr_r18 --weight path/to/weight --show

python demo.py --mode video --path_to_vid path/to/video --cuda -m rtdetr_r18 --weight path/to/weight --show

python demo.py --mode camera --cuda -m rtdetr_r18 --weight path/to/weight --show

Model	Batch	Scale	AP^val 0.5:0.95	AP^val 0.5	FLOPs ^(G)	Params ^(M)	ckpt	Logs
RT-DETR-R18	4xb4	640	45.5	63.5	66.8	21.0	ckpt	log
RT-DETR-R50	4xb4	640	50.6	69.4	112.1	36.7	ckpt	log