PRBonn / PRBonn/PINGS

Poor Results with Custom Data

Open
#13 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
243
Forks
14
PR merge metrics
No merged PRs in 30d

Description

Hi,

I'm trying on a custom dataset with a lidar data without intensity... What are thing i need to take care of in the configs.

  1. Tried a lower batch size of 1024 , 2048 -- seems to help in my case
  2. n_gaussian: simply set to 24 , 48
  3. gaussian n_iters: 1000
  4. Using min and max range values to be 200m and -15 to +40 based on datasheet.
    and what others

Does these make sense. Please let me know @YuePanEdward

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the project's configuration entry points and the documented requirements for custom lidar data. Compare the reported batch size, n_gaussian, gaussian n_iters, and min/max range settings with the expected inputs, including the absence of intensity. Done is a documented, reproducible configuration or a confirmed explanation for the poor results.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.