First of all I want to thank you for your tutorial which really gives me a lot help, I learned a lot through your code.
I am currently working on the Q-learning code, and I see that you used Raw_function instead of actions.FUNCTIONS in the code. I did some research but unfortunately I didn't find the difference between them. I seems to me that the origin actions.FUNCTIONS that you used in those basic agent programs are more similar to real world operations, while the utilisation of those RAW_FUNCTIONS seems to be much easier and a little bit tricky. Am I understanding right ? To simulate human operation in the game, which one should be the better choice ?
Hello!
First of all I want to thank you for your tutorial which really gives me a lot help, I learned a lot through your code.
I am currently working on the Q-learning code, and I see that you used Raw_function instead of actions.FUNCTIONS in the code. I did some research but unfortunately I didn't find the difference between them. I seems to me that the origin actions.FUNCTIONS that you used in those basic agent programs are more similar to real world operations, while the utilisation of those RAW_FUNCTIONS seems to be much easier and a little bit tricky. Am I understanding right ? To simulate human operation in the game, which one should be the better choice ?
Thanks! :)