RuntimeError: Expected isFloatingType(grads[i].type().scalarType()) to be true, but got false.

[jiangtao129]

Dear， A RuntimeError occurred while I was training on the Kitti dataset, as follows

Traceback (most recent call last):
  File "/media/jiangtao/Project/IA-SSD-main/tools/train_utils/train_utils.py", line 52, in train_one_epoch
    loss.backward()

  File "/home/zhang/anaconda3/envs/second1/lib/python3.6/site-packages/torch/tensor.py", line 195, in backward
    torch.autograd.backward(self, gradient, retain_graph, create_graph)

  File "/home/zhang/anaconda3/envs/second1/lib/python3.6/site-packages/torch/autograd/__init__.py", line 99, in backward
    allow_unreachable=True)  # allow_unreachable flag

RuntimeError: Expected isFloatingType(grads[i].type().scalarType()) to be true, but got false.  (Could this error message be improved?  If so, please report an enhancement request to PyTorch.)

Could you please give me the suggestions？ Looking forward to your reply.

my environment Ubuntu18.04 Python = 3.6 PyTorch = 1.4 CUDA = 10.2 spconv = 1.2.1

[strugglingguoguo]

I also met the same problem which let me upset

[itskyf]

Have you modified the code? Mine is working perfectly. But I have found this issue, please check if it could help

[blueskych]

I also met the same problem which let me upset

go to IASSD_head.py, find the function def get_loss(self, tb_dict=None), modfiy all the loss values of point_loss from tensor to scalar.For example, change point_loss_box to point_loss_box.item() in function get_center_box_binori_layer_loss.