Own NLP implementations

[Borda]

[0.5.1] - 2021-08-30

Added

• Added device and dtype properties (Device and dtype properties #462)
• Added TextTester class for robustly testing text metrics (Addition of TextTester class and update tests for current text metrics. #450)

Changed

• Added support for float targets in nDCG metric (Add float target support to class & functional NDCG #437)

Removed

• Removed rouge-score as dependency for text package (Add torchmetrics' own implementation of Rouge score metrics #443)
• Removed jiwer as dependency for text package (Remove dependency on jiwer for WER #446)
• Removed bert-score as dependency for text package (Implement own BERT score #473)

Fixed

• Fixed ranking of samples in SpearmanCorrCoef metric (Fix ranking in spearman metric #448)
• Fixed bug where compositional metrics where unable to sync because of type mismatch (Fix _sync_dist for compositional metrics #454)
• Fixed metric hashing (fixes metric hashing #478)
• Fixed BootStrapper metrics not working on GPU (Device and dtype properties #462)
• Fixed the semantic ordering of kernel height and width in SSIM metric (Change semantic ordering of kernel_size parameter in SSIM #474)

Contributors

@justusschock, @karthikrangasai, @kingyiusuen, @obus, @SkafteNicki, @stancld

If we forgot someone due to not matching commit email with GitHub account, let us know :]

---

This discussion was created from the release Own NLP implementations.