**pred_distri, pred_scores = torch.cat([xi.view(feats[0].shape[0], self.no, -1) for xi in feats]

```python
from ultralytics import YOLO

if __name__ == '__main__':

    model = YOLO(r'./v8\yolov8s.yaml')  # load a pretrained model (recommended for training)
    # model = YOLO(r'C:\dyh\ultralytics\ultralytics-main\ultralytics\cfg\models\v8\yolov8m.yaml')  # load a pretrained model (recommended for training)
    # model = YOLO(r'yolov8s-seg.pt')  # load a pretrained model (recommended for training)
    # model.load(r'C:\dyh\ultralytics\v8_seg_best.pt')  # load a pretrained model (recommended for training)

    # model.load(r"C:\Users\bx\Desktop\dataset\jupi\best.pt")  # load a pretrained model (recommended for training)
    model.load(r"./yolov8s.pt")  # load a pretrained model (recommended for training)

    # model.load(r"C:\dyh\ultralytics\runs\train\v8s_det_1010shujuji_labelimg3\weights\best.pt")  # load a pretrained model (recommended for training)

    model.train(data=r'./coco128-seg-lunpan-biaomiankeng4_mengwei_1010_ganghui_305_labelme_jupi.yaml',
                          epochs=300, imgsz=1024, patience=400,
                          project='runs/train',
                          name='v8s_det_1010shujuji_labelimg1015_jupi',
                          workers=10,
                          batch=6,
                          task='detect',
                          )

Traceback (most recent call last):
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\111_v8_train_detect_1010shujuji.py", line 12, in <module>
    model.train(data=r'C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\coco128-seg-lunpan-biaomiankeng4_mengwei_1010_ganghui_305_labelme_jupi.yaml',
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\engine\model.py", line 601, in train
    self.trainer.train()
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\engine\trainer.py", line 208, in train
    self._do_train(world_size)
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\engine\trainer.py", line 376, in _do_train
    self.loss, self.loss_items = self.model(batch)
  File "C:\Users\bx\AppData\Local\anaconda3\envs\mw\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "C:\Users\bx\AppData\Local\anaconda3\envs\mw\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
    return forward_call(*args, **kwargs)
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\nn\tasks.py", line 82, in forward
    return self.loss(x, *args, **kwargs)
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\nn\tasks.py", line 261, in loss
    return self.criterion(preds, batch)
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\utils\loss.py", line 263, in __call__
    pred_distri, pred_scores = torch.cat([xi.view(feats[0].shape[0], self.no, -1) for xi in feats], 2).split(
  File "C:\Users\bx\Desktop\0904代码\0904代码\ultralytics\ultralytics\utils\loss.py", line 263, in <listcomp>
    **pred_distri, pred_scores = torch.cat([xi.view(feats[0].shape[0], self.no, -1) for xi in feats], 2).split(
RuntimeError: shape '[75, 75, -1]' is invalid for input of size 1920000**

**特征图大小对应不上，看错误定位，进入了分割的loss计算流程，**解决：
**将当前目录下的ultralytics文件夹删除，让代码调用环境中pip安装的ultralytics包而不是使用本地的ultralytics包**