Reza Mahjourian

Can you check the shapes of the input tensors at each stage to see where you start seeing the None shape for the batch dimension? In case you have already...

Does it make a different if you use the same scope around earlier operations as well (when the timestep_grids and waypoints) are being created? In case you have already solved...

Hi, Thank you for your interest in the challenge and for communicating with us. This is a known issue and we are working on resolving it. The results will automatically...

Thank you for your interest. I wouldn't expect the deadline to change. I don't see any new submissions from you after the last one which we evaluated. The evaluation results...

We have evaluated the new submissions from you. You have sent 5 in total. One of them has invalid scenario ids -- probably the test set results were submitted on...

Thank you for your interest in the scenario generation challenge. The rollouts in your submission have only 80 timesteps. This challenge requires 91 timesteps for each trajectory (covering past, current,...

Right. That's something we are working on.

The issue is fixed. Please check your "My Submissions" page and let us know whether or not it's working for you.

Hi, thanks for reporting this issue. Yes, ADV is the same as SDC. Yes, the code should be using `sdc_track_index`. Please note that `logged_trajectories` holds tensors that have been created...

Here is the fix: We have changed the function `get_sim_agent_ids` in `submission_specs.py` to always put the ADV/SDC as the first agent: ``` def get_sim_agent_ids( scenario: scenario_pb2.Scenario, challenge_type: ChallengeType ) ->...