剪枝 python prune.py --percent 0.5 --weights runs/train/exp2/weights/last.pt --data data/voc.yaml --cfg models/yolov5s.yaml出现错误：TypeError: run() got an unexpected keyword argument 'cfg'

[YTTT9]

(yolov5) D:\AAAAAAAAA\YOLOV5-4.0\QQQQ\7Q-yolov5prune-6.0>python prune.py --percent 0.5 --weights weight/last.pt --data data/coco128.yaml yolo: data=data/coco128.yaml, weights=['weight/last.pt'], cfg=D:\AAAAAAAAA\YOLOV5-4.0\QQQQ\7Q-yolov5prune-6.0\models\yolov5s.yaml, percent=0.5, batch_size=8, imgsz=416, conf_thres=0.001, iou_thres=0.6, task=v al, device=, workers=8, single_cls=False, augment=False, verbose=False, save_txt=False, save_hybrid=False, save_conf=False, save_json=False, project=D:\AAAAAAAAA\YOLOV5-4.0\QQQQ\7Q-yolov5prune-6.0\runs\val, n ame=exp, exist_ok=False, half=False, dnn=False test before prune ... Traceback (most recent call last): File "prune.py", line 805, in main(opt) File "prune.py", line 775, in main run(**vars(opt)) File "D:\anaconda\envs\yolov5\lib\site-packages\torch\autograd\grad_mode.py", line 26, in decorate_context return func(*args, **kwargs) TypeError: run() got an unexpected keyword argument 'cfg'

[cxy-ccc]

请问解决了吗

[wynshiter]

在 prune.py 的run 函数中把cfg 参数加上, 下面两个的opt 其实要统一的

if opt.task in ('train', 'val', 'test'):  # run normally
        if opt.conf_thres > 0.001:  # https://github.com/ultralytics/yolov5/issues/1466
            LOGGER.info(f'WARNING: confidence threshold {opt.conf_thres} >> 0.001 will produce invalid mAP values.')
        LOGGER.info(f'test before prune ... ')
        run(**vars(opt))
        LOGGER.info('='*100)
        LOGGER.info('Test after prune ... ')
        run_prune(**vars(opt))