OpenPCSeg issues

train problem

1

Hi Team, I now want to learn how to use the OpenSeg open source toolbox, and I am currently trying to use the Rangnet++ algorithm in the toolbox for training,...

A6Robot

nuscenes

1

Hello, when will the code related to training and testing on nuscenes be published

eheuej

ZeroDivisionError: division by zero

3

Hello! I encountered an error while running the training file: (pcseg) root@9e1fedc6ce4f :/workspace/data/OpenPCSeg master # Python train. py -- cfg_file tools/cfgs/voxel/semantic_kitti/inkunt_mk34_cr10. yaml 2024-04-16 11:44:23277 INFO ********************************** Start logging********************** 2024-04-16 11:44:23277...

ccdontworry

why num_points must always bigger than self.num_points

` _, inds, inverse_map = sparse_quantize(pc_, return_index=True, return_inverse=True) if self.training and len(inds) > self.num_points: # NOTE: num_points must always bigger than self.num_points raise RuntimeError('droping point') inds = np.random.choice(inds, self.num_points, replace=False)`...

Assassincup

Incompatible of optimizer parameters

Hi team, Can you please test the RPVNet pretrained model I imported it and use the same config but there is an error as below: ![image](https://github.com/PJLab-ADG/OpenPCSeg/assets/106338986/8cbd20ac-5335-44fd-aac8-2ea39b442100) The environment is the...

hphnngcquan

How to utilize the pretrained model to infer single-frame point semantic segmentation and visualization

4

LucasYang567

How to test against the semantickitti test dataset

1

I have trained a model and got the weights, how should I test it on the semantickitti test set (sequences 11-21) to get the prediction result file for each point?

Shixiaomeng7