weights=ROOT / ‘yolov5s.pt’, # model.pt path(s)# 权重文件地址 默认 weights/可以是自己的路径!!!
parser.add_argument('--weights', nargs='+', type=str, default=ROOT / r'D:\\project1\\AiTools\\yolov5-6.0\\runs\\train\\whiteKBlittle2\\weights\\best.pt', help='model path(s)')
source=ROOT / ‘data/images’, # file/dir/URL/glob, 0 for webcam 0 自带电脑摄像头, 默认data/images/选择预测的图片的路径!!!
parser.add_argument('--source', type=str, default=ROOT / r'D:\\project1\\AiTools\\data\\A4\\yolodata\\images', help='file/dir/URL/glob, 0 for webcam')
data=ROOT / ‘data/coco128.yaml’, #dataset.yaml path, data文件路径,包括类别/图片/标签等信息!!!
parser.add_argument('--data', type=str, default=ROOT / r'D:\\project1\\AiTools\\yolov5-6.0\\data\\White_KB_A4.yaml', help='(optional) dataset.yaml path')
imgsz=640, # inference size (pixels) 输入图片的大小 默认640*640!!!
parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[256], help='inference size h,w')
conf_thres=0.25, # confidence threshold# object置信度阈值 默认0.25 用在nms中!!!
parser.add_argument('--conf-thres', type=float, default=0.25, help='confidence threshold')
max_det=1000, # maximum detections per image 每张图片最多的目标数量 用在nms中!!!
parser.add_argument('--max-det', type=int, default=30, help='maximum detections per image')
project=ROOT / ‘runs/detect’, # save results to project/name 保存测试日志的文件夹路径!!!
parser.add_argument('--project', default=ROOT / 'runs/detect', help='save results to project/name')
name=‘exp’, # save results to project/name 每次实验的名称!!!
parser.add_argument('--name', default='exp', help='save results to project/name')
line_thickness=3, # bounding box thickness (pixels) 画框的线条粗细!!!
parser.add_argument('--line-thickness', default=1, type=int, help='bounding box thickness (pixels)')
检测框代码如下:
if save_img or save_crop or view_img: # Add bbox to image
c = int(cls) # integer class
label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}')
annotator.box_label(xyxy, label, color=colors(c, True))
if save_crop:
save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True)