Chess Evaluation Board

Github

---

This evaluation board uses an XGBoost AI model to evaluate how good a chess position is based on a few factors.

• The opening that was played
• How good the position is (based on a stockfish evaluation.)
• The rating of the black player
• The rating of the white player

In the example board shown, these factors are automatically resolved and shown.

The percentages are the chances for white to win, black to win and the game to be a draw.

The data to train the model comes directly from the lichess download page.

The dataset of 100,000 games used to train the model can be found here.