luca-g97/Magnus-Carlsen-Lichess-Games-Dataset-FEN

Magnus Carlsen Lichess Games Dataset

Overview

This dataset comprises chess games played by World Chess Champion Magnus Carlsen on Lichess.org. The data has been processed to include detailed information for each move, including the corresponding Forsyth-Edwards Notation (FEN) and a standardized FEN format. This structure facilitates in-depth analysis of Carlsen's gameplay across various positions and scenarios.

Dataset Structure

The dataset is provided in a CSV file named train_data.csv, with the following columns:

• —game: A unique identifier for each game.
• —move_number: The sequential number of the move within the game.
• —event: The event name, date played, and round information.
• —winner: The winner's username, color, and Elo rating.
• —loser: The loser's username, color, and Elo rating.
• —move: The move in UCI (Universal Chess Interface) format.
• —player: The player's username and color for the move.
• —fen: The FEN string representing the board state after the move.
• —standardized_fen: A standardized version of the FEN string.
• —link: A link to the game on Lichess.org.

Data Processing

The dataset was created by processing raw game and move data as follows:

1. 1.Data Acquisition: Games and moves were sourced from the Magnus Carlsen Lichess Games Dataset on Kaggle.

1. 1.Move Validation and FEN Generation: Using the python-chess library, each move was applied to a virtual chessboard to generate the corresponding FEN string. Invalid moves were identified and excluded from the dataset.

1. 1.Data Enrichment: Additional information, such as event details, player usernames, colors, and Elo ratings, was integrated into the dataset.

1. 1.FEN Standardization: The FEN strings were standardized to a consistent format to facilitate analysis.

Usage

This dataset is suitable for various applications, including:

• —Chess Engine Training: Developing and training chess engines to evaluate positions and suggest optimal moves.
• —Game Analysis: Studying Magnus Carlsen's strategies and move choices in different positions.
• —Machine Learning: Applying machine learning techniques to predict moves or outcomes based on board positions.

Acknowledgments

The original data was sourced from the Magnus Carlsen Lichess Games Dataset on Kaggle. The python-chess library was instrumental in processing the moves and generating FEN strings.

License

Please refer to the original dataset's licensing terms for usage guidelines.