search

zhejz / carla-roach

Roach: End-to-End Urban Driving by Imitating a Reinforcement Learning Coach. ICCV 2021.
https://zhejz.github.io/roach
Other
274 stars 50 forks source link

Segmentation fault (core dumped) while training RL expert #22

Closed atg93 closed 1 year ago

atg93 commented 1 year ago

Hi,

First of all thank you very much for sharing your code and your work in the paper is very interesting. But I encounter a problem while training RL expert. I would appreciate if you help me.

I installed carla 0.9.10 like you said in Installation.md. I can run benchmark.py and observe the car's behavior in video log. However, when I run train_rl.py, it shows segmentation fault. Moreover, I also noticed that the problem occurs when self._world tries to get vehicle_bbox_list and walker_bbox_list in chauffeurnet.py.

Have you encountered similar problem while you train your RL expert ?

Thank you,

zhejz commented 1 year ago

No I didn't, but segmentation fault (core dumped) can be caused by many reasons and often it's CARLA related. For example, does your GPU has enough memory? train_rl.py runs pytorch training and multiple CARLA servers on a single GPU.

atg93 commented 1 year ago

Thank you for your answer. I think this problem occurs when carla is installed by using pip. I re-created virtual environment and installed carla by using conda install in virtual environment. Then, the problem is solved.