yjh0410 9c6622cdaa update		před 1 rokem
..
README.md	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
build.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
loss.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
matcher.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
resnet.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
yolov1.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
yolov1_backbone.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
yolov1_basic.py	9c6622cdaa update	před 1 rokem
yolov1_head.py	264112178f build a new YOLO-Tutorial project for my book	před 1 rokem
yolov1_neck.py	87d5a72477 modify init	před 1 rokem
yolov1_pred.py	606c352336 modify pred init	před 1 rokem

Redesigned YOLOv1:

Model	Backbone	Batch	Scale	AP^val 0.5	Weight	Logs
YOLOv1	ResNet-18	1xb16	640	75.0	ckpt	log

For training, we train redesigned YOLOv1 with 150 epochs on COCO.
For data augmentation, we use the SSD's augmentation, including the RandomCrop, RandomDistort, RandomExpand, RandomHFlip and so on.
For optimizer, we use AdamW with weight decay of 0.05 and per image base lr of 0.001 / 64.
For learning rate scheduler, we use cosine decay scheduler.
For batch size, we set it to 16, and we also use the gradient accumulation to approximate batch size of 256.

Train YOLOv1

Taking training YOLOv1-R18 on COCO as the example,

python train.py --cuda -d coco --root path/to/coco -m yolov1_r18 -bs 16 --fp16

Taking training YOLOv1-R18 on COCO as the example,

python -m torch.distributed.run --nproc_per_node=8 train.py --cuda --distributed -d coco --root path/to/coco -m yolov1_r18 -bs 16 --fp16

Taking testing YOLOv1-R18 on COCO-val as the example,

python test.py --cuda -d coco --root path/to/coco -m yolov1_r18 --weight path/to/yolov1.pth --show

Taking evaluating YOLOv1-R18 on COCO-val as the example,

python eval.py --cuda -d coco --root path/to/coco -m yolov1_r18 --weight path/to/yolov1.pth

python demo.py --mode image --path_to_img path/to/image_dirs/ --cuda -m yolov1_r18 --weight path/to/weight --show

python demo.py --mode video --path_to_vid path/to/video --cuda -m yolov1_r18 --weight path/to/weight --show --gif

python demo.py --mode camera --cuda -m yolov1_r18 --weight path/to/weight --show --gif

Model	Backbone	Batch	Scale	AP^val 0.5:0.95	AP^val 0.5	FLOPs ^(G)	Params ^(M)	Weight	Logs
YOLOv1	ResNet-18	1xb16	640	27.6	46.8	37.8	21.3	ckpt	log