Skip to content

Lab 10 Peer Review  #7

Description

@FarisanFekri

I think this code is good. It sets up a Q-learning agent for Tic-Tac-Toe in a clear way using the QAgent class. The use of named tuples for the game state is nice and makes the code easy to follow. The random_game function is handy for creating random game situations during training.

However, I think there are areas to make it even better. The code doesn't handle when the game is finished properly, and it needs a way to recognize that. Adding a function for that would help. Also, using more descriptive names for variables and adding more comments inside the QAgent class methods, especially for the Q-value updates, would make it easier for others to understand. Despite these points, the code is a good starting point for a Tic-Tac-Toe agent using Q-learning.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions