RuntimeError: Output 0 of SplitBackward is a view and is being modified inplace.

[msmekala]

Hello Everone! When i am trying to run python3 train.py ../configs/car_cfg.py

i get the following RunTimeError

"Traceback (most recent call last): File "tools/train.py", line 128, in main() File "tools/train.py", line 118, in main log_interval = cfg.log_config.interval File "/home/rlrc/Downloads/SA-SSD/tools/train_utils/init.py", line 99, in train_model log_interval = log_interval File "/home/rlrc/Downloads/SA-SSD/tools/train_utils/init.py", line 57, in train_one_epoch outputs = batch_processor(model, data_batch) File "/home/rlrc/Downloads/SA-SSD/tools/train_utils/init.py", line 29, in batch_processor losses = model(data) File "/home/rlrc/anaconda3/envs/pcdet/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl return forward_call(*input, *kwargs) File "/home/rlrc/Downloads/SA-SSD/mmcv/mmcv/parallel/data_parallel.py", line 50, in forward return super().forward(inputs, kwargs) File "/home/rlrc/anaconda3/envs/pcdet/lib/python3.7/site-packages/torch/nn/parallel/data_parallel.py", line 166, in forward return self.module(*inputs[0], kwargs[0]) File "/home/rlrc/anaconda3/envs/pcdet/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl return forward_call(input, kwargs) File "/home/rlrc/Downloads/SA-SSD/mmdet/models/detectors/base.py", line 79, in forward return self.forward_train(img, img_meta, kwargs) File "/home/rlrc/Downloads/SA-SSD/mmdet/models/detectors/single_stage.py", line 103, in forward_train bbox_score = self.extra_head(conv6, guided_anchors) File "/home/rlrc/anaconda3/envs/pcdet/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl return forward_call(input, kwargs) File "/home/rlrc/Downloads/SA-SSD/mmdet/models/single_stage_heads/ssd_rotate_head.py", line 439, in forward (xs, ys) = self.gen_grid_fn(ga[:, [0, 1, 3, 4, 6]]) File "/home/rlrc/Downloads/SA-SSD/mmdet/models/single_stage_heads/ssd_rotate_head.py", line 380, in gen_samplegrid xg = xg.unsqueeze(-1).expand(N, *window_size) RuntimeError: Output 0 of SplitBackward is a view and is being modified inplace. This view is an output of a function that returns multiple views. Inplace operators on such views is forbidden. You should replace the inplace operation by an out-of-place one."

My Torch version is 1.9.0+cu102

Thanks a lot for your help and support in advance

[sylivahf]

1.9.0+cu102

Thanks a lot for your help and support in

Hi，did you solve it? Thanks a lot.

[AOOOOOA]

I had the same problem with rtx3090+cuda11.1+torch1.7.0+spconv 2.2(also has error with spconv 1.0)\

How do you solve the problem? Thanks

[Chenghao-Wen]

I managed to solve this problem by modifying the code as following:

xg1, yg1, wg1, lg1, rg1 = torch.split(box, 1, dim=-1) xg=xg1.clone() yg=yg1.clone() wg=wg1.clone() lg=lg1.clone() rg=rg1.clone()