Hello, I have a question. In the code, you mentioned that the input to the WaypointPlanner module includes the occupancy result from the perception module. However, I couldn't find the code segment where you generate the occupancy prediction based on the perception results. Instead, I saw in the pnp_dataset.py file that you directly used the ground truth from the dataset to generate the occupancy and input it into the prediction module. I would like to know how the occupancy is actually generated.
