https://pypi.org/project/imbalanced-losses/ Recent updates to the Python Package Index for imbalanced-losses en https://pypi.org/project/imbalanced-losses/0.3.1/ 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. cjsantiago@gatech.edu Wed, 29 Apr 2026 21:21:10 GMT https://pypi.org/project/imbalanced-losses/0.3.0/ 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. cjsantiago@gatech.edu Wed, 29 Apr 2026 12:56:43 GMT https://pypi.org/project/imbalanced-losses/0.2.3/ 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. cjsantiago@gatech.edu Wed, 29 Apr 2026 03:30:13 GMT https://pypi.org/project/imbalanced-losses/0.2.2/ 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. cjsantiago@gatech.edu Sun, 22 Mar 2026 18:33:25 GMT https://pypi.org/project/imbalanced-losses/0.2.1/ 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. cjsantiago@gatech.edu Sun, 22 Mar 2026 18:03:10 GMT https://pypi.org/project/imbalanced-losses/0.2.0/ 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. cjsantiago@gatech.edu Sat, 14 Mar 2026 19:12:54 GMT