tomascanivari/mask2former-swin-large-coco-instance-finetuned-buildings

Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. --> This model was used in the context of https://www.kaggle.com/competitions/building-extraction-generalization-2024/overview challenge (after ending).

Trained on the tomascanivari/buildings-extract-coco-hf dataset for 12 epochs.

On the test set:

• —Achieved around 0.85 pixel-wise values in the metrics: precision, recall, f1-score and accuracy.

• —Achieved around 0.60 object-wise f1-score (target metric).