yolov8-pose在转wts时，报错AttributeError: 'NoneType' object has no attribute 'float'

[wllzq126]

配置：
NVIDIA GeForce RTX 3090 CUDA Version: 12.4
nvidia-cublas-cu12 12.1.3.1
nvidia-cuda-cupti-cu12 12.1.105
nvidia-cuda-nvrtc-cu12 12.1.105
nvidia-cuda-runtime-cu12 12.1.105
nvidia-cudnn-cu12 8.9.2.26
nvidia-cufft-cu12 11.0.2.54
nvidia-curand-cu12 10.3.2.106
nvidia-cusolver-cu12 11.4.5.107
nvidia-cusparse-cu12 12.1.0.106
nvidia-nccl-cu12 2.20.5
nvidia-nvjitlink-cu12 12.6.85
nvidia-nvtx-cu12 12.1.105
opencv-python 4.10.0.84
torch 1.12.1+cu113
torchvision 0.13.1+cu113
ultralytics 8.2.103

我在运行python gen_wts.py -w "/mnt/storage/wll/data/pose/backup/yolov8/train7/weights/best.pt" -o "/mnt/storage/wll/data/pose/backup/yolov8/train7/weights/best.wts" -t pose
这个命令式，报错信息如下：
Loading /mnt/storage/wll/data/pose/backup/yolov8/train7/weights/best.pt
Traceback (most recent call last):
File "gen_wts.py", line 40, in
model = torch.load(pt_file, map_location=device)['model'].float() # load to FP32
AttributeError: 'NoneType' object has no attribute 'float'

然而，我在转换detect模型文件是，是可以正常转换的。
请问这个问题应该怎么解决？非常感谢

[mpj1234]

你用的pt模型是没有训练结束的把吧，如果是没有训练结束的模型还保存在ckpt的ema属性上面，所以你的ckpt的model拿不到东西