yfzhang/vehicle-motion-forecasting

A PyTorch-based deep inverse reinforcement learning pipeline for vehicle motion forecasting

vehicle-motion-forecasting's Issues

Will data or sample data be available?

Rotations

In the paper, you say that "augmenting the demonstration data with different rotations is critical because the proposed two-stage network is sensitive to directional bias."

As there is nothing in the code that performs rotations I assume you do this beforehand externally on the dataset and not 'live' during training.

My question is regarding how you rotate the trajectories along-with the environmental grids. As we have an even-numbered grid size (80x80) surely rotating the trajectories will result in the rotated trajectories having a different start state from the [40,40] state? Is this a problem? Is it important that all future trajectories start from [40,40]? If so is there a way to rotate the trajectories to ensure this start state remains constant?

How did you apply the train of model of reward

Hay

Thank you very much for your work

I have question how did you train and model of reward using the equation :

how did you calculte the state visitation frequency from expert data and learner

Sincerly

dataset

Can you provide a dataset or a step-by-step method for processing the dataset? thanks