dsvilarkovic / dynamic_plane_convolutional_onet Goto Github PK

Hi

I'm trying to reproduce the results of this repo. I have installed the conda environment along with dependencies, pretrained models and dataset correctly. I'm trying to perform the "Mesh Generation" step using pre-trained models. By using the following command

python generate.py configs/pointcloud/shapenet/shapenet_dynamic_3plane.yaml

I'm getting this Error. Can you help me solving it? System Information : I'm running this code on a Tesla T4 GPU.

<src.data.core.Shapes3dDataset object at 0x7f16ae654048>
Traceback (most recent call last):
  File "generate.py", line 45, in <module>
    checkpoint_io.load(cfg['test']['model_file'])
  File "/home/ubuntu/dynamic_plane_convolutional_onet/src/checkpoints.py", line 49, in load
    return self.load_file(filename)
  File "/home/ubuntu/dynamic_plane_convolutional_onet/src/checkpoints.py", line 68, in load_file
    raise FileExistsError
FileExistsError
Segmentation fault (core dumped)

Hi,

Thank you for releasing the code. It has been very helpful while reading the paper to clarify the theory in the implementation. In the Installation section and Dataset sub-section, under the ShapeNet and Synthetic Room paragraphs, the following commands has been mentioned

bash src/scripts/download_shape_net_data.sh
bash src/scripts/download_room_data.sh

While running this code, I found out that for this repository the files are located in the "scripts" and not "src/scripts" . So the commands that worked for me are

bash scripts/download_shape_net_data.sh
bash scripts/download_room_data.sh

Please let me know if I'm missing something. Thanks!

How much resolution is used to reconstruction the synthetic rooms dataset in your article? 128^3 or 256^3?