ONNX
English
YOLO
ONNX
onnxruntime
balatro

Balatro card edition detection

YOLO11n detector for none, foil, holographic, polychrome, and negative editions. This model is part of the Project AIRI game-playing work.

Files

  • model.pt: Ultralytics/PyTorch checkpoint.
  • onnx/model.onnx: fixed-shape ONNX export at 640 px.
  • docs/: training curves, confusion matrices, and prediction examples.

Independent test split

Metric Value
Precision 0.998739
Recall 0.999392
mAP50 0.995000

The train/validation/test split is grouped by generated game round. Metrics describe the Mod-generated distribution and are not a claim of identical accuracy on arbitrary live captures.

Prediction examples

Batch 0 Batch 1 Batch 2

Training diagnostics

License

MIT.

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. 🙋 Ask for provider support

Model tree for proj-airi/games-balatro-2024-yolo-card-edition-detection-mod-ground-truth

Quantized
(88)
this model