CUDA error

[827346462]

ghts ./weights/helmet_head_person_m.pt
Namespace(agnostic_nms=False, augment=False, classes=None, conf_thres=0.4, device='', img_size=640, iou_thres=0.5, output='inference/output', save_txt=False, source='000017.jpg', update=False, view_img=False, weights=['./weights/helmet_head_person_m.pt'])
Using CUDA device0 _CudaDeviceProperties(name='NVIDIA GeForce RTX 3090', total_memory=24267MB)

Fusing layers... Traceback (most recent call last):
File "detect.py", line 161, in
detect()
File "detect.py", line 23, in detect
model = attempt_load(weights, map_location=device) # load FP32 model
File "/home/sevnce/Smart_Construction/Smart_Construction-master/models/experimental.py", line 133, in attempt_load
model.append(torch.load(w, map_location=map_location)['model'].float().fuse().eval()) # load FP32 model
File "/home/sevnce/Smart_Construction/Smart_Construction-master/models/yolo.py", line 150, in fuse
m.conv = torch_utils.fuse_conv_and_bn(m.conv, m.bn) # update conv
File "/home/sevnce/Smart_Construction/Smart_Construction-master/utils/torch_utils.py", line 113, in fuse_conv_and_bn
fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.size()))
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)

我在ubuntu 18.04 cuda11.2 rtx3090 环境 虚拟环境是requirment.txt.
直接运行 python detect.py --source 000017.jpg --weights ./weights/helmet_head_person_m.pt
出现cuda 错误。
请问这个cuda版本和权重不适配的问题吗