Welcome to the ConnectX Simulation Competition repository! This project is part of a beta-version simulation competition hosted on Kaggle, where participants compete against a set of rules using Python submissions.
This competition challenges participants to develop an AI agent capable of playing the game ConnectX. The objective is to connect a certain number of checkers in a row—horizontally, vertically, or diagonally—before the opponent does. Participants must submit a Python .py
file that acts as an AI agent to play against other submissions.
- Game Objective: Achieve a specified number of checkers in a row before the opponent on a game board.
- Submission: Participants submit a Python
.py
file containing their AI agent's logic. - Evaluation: Submissions are evaluated based on their performance against other submissions, rather than a traditional accuracy metric.
- Rating System: Uses a Gaussian model to estimate skill levels of submissions.
To participate in this competition:
-
Clone the Repository: Clone this repository to your local machine.
git clone https://github.com/your-username/connectx-competition.git
-
Set Up Your Environment: Ensure you have Python installed. You may also need to install the
kaggle-environments
package.pip install kaggle-environments
-
Develop Your Agent: Modify the
submission.py
file to implement your AI agent. Example starter code is provided to get you started. -
Submit Your Agent: Once you're satisfied with your agent's performance locally, submit your
submission.py
to Kaggle through their competition interface.
submission.py
: Main file where your AI agent's logic is implemented. Modify this file to improve your agent's performance.README.md
: This file provides an overview of the competition, setup instructions, and guidelines for participating.
- Your agent must return an action within 2 seconds (60 seconds on the first turn) of being invoked.
- Use only modules from the Kaggle Kernels notebook image.
- Ensure your submission does not exceed the maximum file size limit of 100 MB.
This competition is a beta launch, and your feedback is valuable. If you encounter issues or have suggestions for improvements, please open an issue or pull request. We appreciate your input!