Skip to content
This repository was archived by the owner on Dec 31, 2024. It is now read-only.

Evaluation Code #1

Open
haomengz opened this issue Nov 4, 2023 · 4 comments
Open

Evaluation Code #1

haomengz opened this issue Nov 4, 2023 · 4 comments

Comments

@haomengz
Copy link

haomengz commented Nov 4, 2023

Hi,

Thanks for the nice work! I am interested in reproducing the results, but the evaluation procedure in the repo is not very clear. It seems that we could not directly use the linked ST-P3 eval code. Could you provide the script or more details on how you conduct the evaluate on NuScenes dataset? Thanks!

@abhigoku10
Copy link

abhigoku10 commented Nov 13, 2023

@PointsCoder i am having the same issue here can u point or elaborate the steps to do the evaluation on nuscenes or custom dataset
THanks in advance

@haomengz
Copy link
Author

For the collision evaluation, ST-P3 is using segmentations for calculation. What segmentation results are you using for the calculation? Did you follow the ST-P3 pipeline, or using the data from UniAD? Thanks!

@Gear420
Copy link

Gear420 commented Dec 4, 2023

same issue; .pkl cannot be directly applied to the evaluation code of st-p3.

@PointsCoder
Copy link
Owner

@abhigoku10 @haomengz @Gear420 We have released the code and data for evaluation here.

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

4 participants