How to run the code? about multi-behavior-learning-for-socially-compatible-autonomous-driving

[Sanjulavj]

Hi, the scripts folder exists in the repository at https://github.com/Sanjulavj/Multi-behavior-Learning-for-Socially-Compatible-Autonomous-Driving/tree/main/DRL/scripts.
Please update your local code path accordingly.

[kongxincaizi]

Hi, the scripts folder exists in the repository at https://github.com/Sanjulavj/Multi-behavior-Learning-for-Socially-Compatible-Autonomous-Driving/tree/main/DRL/scripts. Please update your local code path accordingly.

Hello, I'm very sorry. Last time after receiving your reply, I was eager to learn this code and forgot to say thanks.

1. During this period, I have been studying your code. I met the problem about the weights_arr = np.array(
   [1.753971587, -0.5295225956, -0.5097452474, -3.058776862, -1.799733349, -1.1420489, -9.950304272,
   -3.498063628]). This value was obtained through IRL? I couldn't find the corresponding values in the testing-log_80~2284 files under the IRL file. I also checked the average value of the 'weights' in this files, but it doesn't match. So, how did you obtain the weights_arr? Did you not put the corresponding testing-log file in the github repository?

2. Then, I would like to know how you trained the DRL by the random selection of SVO, how is it carried out? I noticed that you have annotated the code in the sb3_highway.py "# model.learn(int(25e3), callback = checkpoint_callback)", and the code in the intersection_env.py "if self.vehicle._random_number == -0.45:.........", but I did not find corresponding code for random selection of SVO. So how to write this code?

Thank you again sincerely! And I'm very looking forward to your reply.
Best wishes!