AllenXiang

Hi, it seems that you have compiled the pointnet2_ops extension with cpu version PyTorch. `RuntimeError: false INTERNAL ASSERT FAILED at "pointnet2_ops/_ext-src/src/sampling.cpp":83, please report a bug to PyTorch. CPU not supported`...

Perhaps some of the tensors or operations are on the cpu. Please try manually specify the gpus before runing the training script as follows: `export CUDA_VISIBLE_DEVICES='2'` `python train.py --config ./configs/pcn_cd1.yaml`

Hi, You can use the Imdb reader of [GRNet](https://github.com/hzxie/GRNet/blob/master/utils/lmdb_serializer.py) to get the .pcd files from the [PCN](https://drive.google.com/drive/folders/1P_W1tz5Q4ZLapUifuOE4rFAZp6L1XTJz) dataset.

Hi, the current test.py file does not store the generated point clouds. You may need to manually save the generated [point clouds](https://github.com/AllenXiangX/SnowflakeNet/blob/0880ba1e1bf1c135d276aa0044f0acd38c44d94e/completion/test.py#L71C19-L71C19) according to your own need.

Hi, you can refer to [PCN](https://github.com/wentaoyuan/pcn) to generate more classes from KITTI dataset.

Hi, sorry for the late response! We get the ground truth point clouds (8192 points) with Poisson disk sampling from the [testing meshes](https://drive.google.com/file/d/1BNqjidBVWP0_MUdMTeGy1wZiR6fqyGmC/view) provided by PU-GAN.

Hi, Thank you for your interest in our work. To have a fair comparisn with previous work on the PCN benchmark, the model is trained on the training set and...