chessai-python 0.0.2

Description:

chessaipython 0.0.2

ChessAI - Chinese Chess Game Analyzer
ChessAI is a groundbreaking tool that brings together computer vision, chess algorithms, and advanced analytics to revolutionize the Chinese Chess analytics landscape. With ChessAI, you don't need expensive electronic boards to analyze your games. Simply use your regular board, set up a camera to capture the position, and let ChessAI do the rest.

Main source code: chesssai.
Deep Learning / Data Preparation: dnn_models/data_preparation - Currenly only support for Chinese Chess (XiangQi), contact me for the license and the source code of the data preparation tool.
Deep Learning / Training: dnn_models/training.


Roadmap

Chess position detection.
Chess engine integration.
Move suggestion.
Deep learning model for chess board detection (No need to use ARUCO markers).

Environment setup

Requirements: Python 3.9, Conda, Node.js 18+.
Clone this repository.

git clone https://github.com/vietanhdev/chessai --recursive


Create a new conda environment and the required packages.

conda create -n chessai python=3.9
conda activate chessai
pip install -e .


Install Node.js packages and build the frontend.

cd chessai/frontend
npm install
cd ..
bash build_frontend.sh

Build chess engine

This project uses godogpaw as the chess engine.
Install Go.
Build the engine.

cd godogpaw
go build


Copy the executable file (godogpaw*) to the ./data/engines folder.

Run the app
ENGINE_PATH="data/engines/godogpaw-macos-arm" python -m chessai.app --run_app

Replace ENGINE_PATH with the path to the chess engine executable file.
Data preparation & Training
This project uses computer vision and deep learning to detect chess pieces and chess board position.
AI flow for chess position detection:


Go to dnn_models folder and follow the instructions in the README.md file to prepare the data and train the model.

References

This project was initially built for Hackster's OpenCV AI Competition 2023. Hackster Project: ChessAI - Chinese Chess Game Analyzer.
Object detection model (for chess pieces) is based on YOLOX - License: Apache 2.0.
Chess engine: godogpaw - License: MIT.
UI components: shadcn-ui - License: MIT.