Pytorch-metric-learning

Features

For some loss functions, labels are now optional if `indices_tuple` is provided:
python
loss = loss_func(embeddings, indices_tuple=pairs)


The losses for which you can do this are:

- CircleLoss
- ContrastiveLoss
- IntraPairVarianceLoss
- GeneralizedLiftedStructureLoss
- LiftedStructureLoss
- MarginLoss
- MultiSimilarityLoss
- NTXentLoss
- SignalToNoiseRatioContrastiveLoss
- SupConLoss
- TripletMarginLoss
- TupletMarginLoss

This issue has come up several times:

412
490
482
473
179
263

New features

- Added [InstanceLoss](https://kevinmusgrave.github.io/pytorch-metric-learning/losses/#instanceloss). See 410 by layumi

Bug fixes

- Fixed a bug in BatchEasyHardMiner where `get_max_per_row` was not always returning correct values, resulting in invalid pairs and triplets. 476

Bug fixes

- Fixed ThresholdReducer being incompatible with older versions of PyTorch (465)
- Fixed VICRegLoss being incompatible with older versions of PyTorch, and missing a division by 2 (467 and 470 by cwkeam)

Other

- Made CustomKNN more memory efficient by removing `torch.cat` call.

Features

- Added a batch_size parameter to [CustomKNN](https://kevinmusgrave.github.io/pytorch-metric-learning/inference_models/#customknn). This computes k-nn per batch of query embeddings (using [BatchedDistance](https://kevinmusgrave.github.io/pytorch-metric-learning/distances/#batcheddistance)), which requires less memory than computing the entire distance matrix at once.

Bug Fixes

- 451 (thanks cwkeam)
- 453

Bug Fixes

- 447