Penghao Wu

We are uploading the dataset to baidu cloud, and the download link will be released soon.

Hi, sorry for the delay. You can now download through the link https://pan.baidu.com/s/11xBZwAWQ3WxQXecuuPoexQ with code 8174.

Hi, do you set up the pre-built Carla as instructed in the setup section? In our case, we do not need to reboot the computer.

1. We use the feature-level supervision from the roach expert which is a learned model. And we use its predicted control distribution parameters for control action prediction. You can refer...

Since there are some small modifications, we will re-train the model for our validation experiments and release it.

Hi, you should set the root_dir_all variable in config.py to the actual directory of your data.

Sorry about this problem. Those files might get ignored for the Baidu Cloud zip file. You can use ```tools/gen_data.py``` to generate packed_data.npy files.

Could you please check the content of the ```packed_data.npy``` to see if it is correctly generated?

Could you please provide all the error information? Usually, this del leaderboard_evaluator error happens after some other exceptions or errors.

I do not remember the exact routes of them, but they should come from the Town05-long and Town02-long routes in the [Transfuser](https://github.com/autonomousvision/transfuser/tree/cvpr2021/leaderboard/data) repo.