dbflip

confirmedupstream bugPython · torchmetricsInference & servingJuly 24, 2026

the question

Does torchmetrics' DaviesBouldinScore declare the correct optimization direction for MetricTracker?

what came out

torchmetrics declares DaviesBouldinScore.higher_is_better=True, but the Davies-Bouldin index is lower-is-better (min 0). MetricTracker reads the flag and maximizes it, selecting the worst clustering (DB 5.94) as best over the good one (0.026). The flag and docstring were copied from Calinski-Harabasz. Fix: higher_is_better=False.

method & receipts

Result: confirmed
Reproducible — one script re-runs the whole thing from scratch.
Tested — a correctness/benchmark suite ships alongside the code.

→ read the code and re-run it

github.com/v-code01/dbflip