imbalanced-losses is a PyTorch library of training losses for class-imbalanced classification — including Focal Loss, Smooth-AP, and Recall-at-Quantile — with built-in DDP all-gather support for globally-correct rank estimation and normalization across multi-GPU training.
An exporter that collects Xray (and V2Ray) metrics over its Stats API and exports them to Prometheus. It also provides enhanced user activity metrics by parsing access logs.