RuntimeError: CUDA 错误: 在调用 cublasSgemv(handle, op, m, n, &alpha, a, lda, x, incx, &beta, y, incy) 时,出现 CUBLAS_STATUS_INVALID_VALUE

当我使用命令 model = YOLOv10('yolov10n.yaml') 选择模型后，在训练 yolov10n.yaml 时，完成一个 epoch 循环训练后进行验证时，会出现 CUDA 错误: CUBLAS_STATUS_INVALID_VALUE when calling `cublasSgemv(handle, op, m, n, &alpha, a, lda, x, incx, &beta, y, incy`。并伴随有警告 UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor'). index = index // nc。但是当我训练 yolov10x.yaml 时，会产生警告，但没有出现错误。请问这是什么原因？如何修复？