When one photo carries several labels
A photos × labels grid where ground truth is fixed and you click the model's predictions. Subset accuracy, Hamming loss and micro/macro F1 update together, which makes the gap between them obvious.
Try this
Get one label wrong on one photo. Hamming loss moves a little; subset accuracy loses the whole photo.
Dot = the true tag. Click a cell to toggle the model's prediction.
Where this lab comes from
Multiclass and Multilabel Metrics: Micro, Macro, Weighted, and the Difference That MattersHow precision/recall/F1 scale past two classes: one-vs-rest, and the micro/macro/weighted averaging that decides whether rare classes count. Plus multilabel-specific metrics: subset accuracy, Hamming loss, and top-k, all on one running shelter-camera example, with scikit-learn.
More in metrics
- Precision, recall and the threshold — Edit the four counts, or drag a threshold and trace the PR and ROC curves.
- IoU and Dice, by dragging boxes — Two draggable boxes; intersection, union and both scores update live.
- Macro, micro and weighted, side by side — An editable C×C confusion matrix with per-class F1 drawn as bars.