YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Senchess AI Models

Collection of YOLOv8 models for chess piece detection.

πŸ“¦ Available Models

πŸ‰ Kaido v1.0 (kaido_v1.0.pt) - RECOMMENDED

  • Best overall performance on styled and standard boards
  • 32/32 pieces detection with 98.61% confidence
  • Trained on 1,000 positions, 50 epochs
  • mAP50: 99.95% | Precision: 99.5% | Recall: 99.5%
  • Size: 6.25 MB
  • Use case: General purpose, excellent for all board styles

πŸ”± Haki v1.0 (haki_v1.0.pt)

  • Specialized for strategic pieces (King, Queen, Rook, Bishop)
  • 32/32 pieces detection with 98.52% confidence
  • Trained on styled boards
  • Size: 6.25 MB
  • Use case: Styled/artistic chess boards

βš™οΈ Gear v1.1 (gear_v1.1.pt)

  • Optimized for standard wooden boards
  • mAP50: 98.79%
  • Size: 6.25 MB
  • Use case: Classic chess boards only
  • ⚠️ May fail on styled boards

πŸš€ Usage

πŸ“Š Performance Comparison

Model Pieces Detected Confidence mAP50 Best For
πŸ‰ Kaido v1.0 32/32 98.61% 99.95% All boards ⭐
πŸ”± Haki v1.0 32/32 98.52% - Styled boards
βš™οΈ Gear v1.1 0-32 varies 98.79% Standard boards

🎯 Classes (12)

  • white-king, white-queen, white-rook, white-bishop, white-knight, white-pawn
  • black-king, black-queen, black-rook, black-bishop, black-knight, black-pawn

πŸ“ License

MIT

οΏ½οΏ½ Repository

https://github.com/MedouneSGB/Senchess-AI

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support