PWhiddy / PokemonRedExperiments

Playing Pokemon Red with Reinforcement Learning
MIT License
6.88k stars 629 forks source link

Introduces new classes for maintainability ✂️ #170

Open mdreano opened 7 months ago

mdreano commented 7 months ago

First of all, thank you for sharing this repository it has been a really helpful introduction for me ❤️

Extracted some logic in separate classes:

Dissociate updating rewards from reading rewards. This is easily seen in the step function of the gym.

Added "Stream" in config Allows to tell if you want to stream to the shared map.

Added "seen_pokemon" reward

Added the current location in the print_reward

Moved sessions in a "session" folder