Skip to content
This repository has been archived by the owner on Nov 20, 2023. It is now read-only.

prediction skipped due to error #82

Open
arxidinakbar opened this issue Jun 9, 2020 · 2 comments
Open

prediction skipped due to error #82

arxidinakbar opened this issue Jun 9, 2020 · 2 comments

Comments

@arxidinakbar
Copy link

Did anyone meet with this issue when running the train.py? I follow every step according the tutorial. Details here below.

predict ['005988' '006849']
/home/omnisky/voxelnet/utils/utils.py:677: RuntimeWarning: overflow encountered in exp
deltas[..., [3, 4, 5]]) * anchors_reshaped[..., [3, 4, 5]]
prediction skipped due to error

@MinghuiJ
Copy link

I get the same error, anyone has some idea?

@arxidinakbar
Copy link
Author

@MinghuiJ Here's some breakthrough with this issue. I just rearrange my dataset constructure according the tutorial again, and running the crop.py, then split the training dataset into training & validation dataset( make sure running the crop.py first, then split), suddenly it works. maybe it's go wrong with the dataset constructure, hope it's helpful

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants