NVlabs / ScePT

Code for the CVPR 2022 paper "ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning" by Yuxiao Chen, Boris Ivanovic, and Marco Pavone
Other
81 stars 17 forks source link

Planner Integration #14

Open ImmerHenry opened 2 weeks ago

ImmerHenry commented 2 weeks ago

Hi,

thank you for your amazing work.

I read from your paper that you also integrated a downstream planner to show the planner outputs with your ScePT prediction.

I find you have planner classes in Planning directory, but there is no use cases in your repo. Are there any examples of how to integrate ScePT outputs with the Planner you mentioned in your paper?

Btw, in README, you mentioned to run train_clique.py for training your model on nuscenes dataset, but there is no such file in your repo. Is it just the same as train.py?

Thank you in advance for your help!

Best, Henry