diff --git a/CHANGELOG.md b/CHANGELOG.md index c70841a3aed..631eb0f6efb 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -7,100 +7,58 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 **Note: we move fast, but still we preserve 0.1 version (one feature release) back compatibility.** -## [unreleased] - 2021-MM-DD +## [0.5.0] - 2021-08-09 ### Added - Added **Text-related (NLP) metrics**: - - Word Error Rate (WER) ([#52](https://github.com/PyTorchLightning/metrics/issues/52)) - - - ROUGE ([#399](https://github.com/PyTorchLightning/metrics/issues/399)) - - - BERT score ([#365](https://github.com/PyTorchLightning/metrics/issues/365)) - - + - Word Error Rate (WER) ([#383](https://github.com/PyTorchLightning/metrics/pull/383)) + - ROUGE ([#399](https://github.com/PyTorchLightning/metrics/pull/399)) + - BERT score ([#424](https://github.com/PyTorchLightning/metrics/pull/424)) + - BLUE score ([#360](https://github.com/PyTorchLightning/metrics/pull/360)) - Added `MetricTracker` wrapper metric for keeping track of the same metric over multiple epochs ([#238](https://github.com/PyTorchLightning/metrics/pull/238)) - - - Added other metrics: - - Symmetric Mean Absolute Percentage error (SMAPE) ([#375](https://github.com/PyTorchLightning/metrics/issues/375)) - - - Calibration error ([#394](https://github.com/PyTorchLightning/metrics/issues/394)) - - - Permutation Invariant Training (PIT) ([#294](https://github.com/PyTorchLightning/metrics/issues/294)) - - -- Added support in `nDCG` metric for target with values larger than 1 ([#343](https://github.com/PyTorchLightning/metrics/issues/343)) - - + - Symmetric Mean Absolute Percentage error (SMAPE) ([#375](https://github.com/PyTorchLightning/metrics/pull/375)) + - Calibration error ([#394](https://github.com/PyTorchLightning/metrics/pull/394)) + - Permutation Invariant Training (PIT) ([#384](https://github.com/PyTorchLightning/metrics/pull/384)) +- Added support in `nDCG` metric for target with values larger than 1 ([#349](https://github.com/PyTorchLightning/metrics/pull/349)) - Added support for negative targets in `nDCG` metric ([#378](https://github.com/PyTorchLightning/metrics/pull/378)) - - - Added `None` as reduction option in `CosineSimilarity` metric ([#400](https://github.com/PyTorchLightning/metrics/pull/400)) - - -- Allowed passing labels in (n_samples, n_classes) to `AveragePrecision` ([#386](https://github.com/PyTorchLightning/metrics/issues/386)) - +- Allowed passing labels in (n_samples, n_classes) to `AveragePrecision` ([#386](https://github.com/PyTorchLightning/metrics/pull/386)) ### Changed - Moved `psnr` and `ssim` from `functional.regression.*` to `functional.image.*` ([#382](https://github.com/PyTorchLightning/metrics/pull/382)) - - - Moved `image_gradient` from `functional.image_gradients` to `functional.image.gradients` ([#381](https://github.com/PyTorchLightning/metrics/pull/381)) - - - Moved `R2Score` from `regression.r2score` to `regression.r2` ([#371](https://github.com/PyTorchLightning/metrics/pull/371)) - - - Pearson metric now only store 6 statistics instead of all predictions and targets ([#380](https://github.com/PyTorchLightning/metrics/pull/380)) - - - Use `torch.argmax` instead of `torch.topk` when `k=1` for better performance ([#419](https://github.com/PyTorchLightning/metrics/pull/419)) - - - Moved check for number of samples in R2 score to support single sample updating ([#426](https://github.com/PyTorchLightning/metrics/pull/426)) - ### Deprecated - Rename `r2score` >> `r2_score` and `kldivergence` >> `kl_divergence` in `functional` ([#371](https://github.com/PyTorchLightning/metrics/pull/371)) - - - Moved `bleu_score` from `functional.nlp` to `functional.text.bleu` ([#360](https://github.com/PyTorchLightning/metrics/pull/360)) - ### Removed - Removed restriction that `threshold` has to be in (0,1) range to support logit input ( [#351](https://github.com/PyTorchLightning/metrics/pull/351) [#401](https://github.com/PyTorchLightning/metrics/pull/401)) - - - Removed restriction that `preds` could not be bigger than `num_classes` to support logit input ([#357](https://github.com/PyTorchLightning/metrics/pull/357)) - - - Removed module `regression.psnr` and `regression.ssim` ([#382](https://github.com/PyTorchLightning/metrics/pull/382)): - - - Removed ([#379](https://github.com/PyTorchLightning/metrics/pull/379)): * function `functional.mean_relative_error` * `num_thresholds` argument in `BinnedPrecisionRecallCurve` - ### Fixed - Fixed bug where classification metrics with `average='macro'` would lead to wrong result if a class was missing ([#303](https://github.com/PyTorchLightning/metrics/pull/303)) - - -- Fixed `weighted`, `multi-class` AUROC computation to allow for 0 observations of some class, as contribution to final AUROC is 0 ([#348](https://github.com/PyTorchLightning/metrics/issues/348)) - - +- Fixed `weighted`, `multi-class` AUROC computation to allow for 0 observations of some class, as contribution to final AUROC is 0 ([#376](https://github.com/PyTorchLightning/metrics/pull/376)) - Fixed that `_forward_cache` and `_computed` attributes are also moved to the correct device if metric is moved ([#413](https://github.com/PyTorchLightning/metrics/pull/413)) - - - Fixed calculation in `IoU` metric when using `ignore_index` argument ([#328](https://github.com/PyTorchLightning/metrics/pull/328)) + ## [0.4.1] - 2021-07-05 ### Changed @@ -150,7 +108,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ### Deprecated -- Deprecated `functional.mean_relative_error` ([#248](https://github.com/PyTorchLightning/metrics/pull/248)) +- Deprecated `functional.mean_relative_error`, use `functional.mean_absolute_percentage_error` ([#248](https://github.com/PyTorchLightning/metrics/pull/248)) - Deprecated `num_thresholds` argument in `BinnedPrecisionRecallCurve` ([#322](https://github.com/PyTorchLightning/metrics/pull/322)) ### Removed diff --git a/torchmetrics/__about__.py b/torchmetrics/__about__.py index 9137e7eaefd..b172f3ca95a 100644 --- a/torchmetrics/__about__.py +++ b/torchmetrics/__about__.py @@ -1,4 +1,4 @@ -__version__ = "0.5.0rc0" +__version__ = "0.5.0" __author__ = "PyTorchLightning et al." __author_email__ = "name@pytorchlightning.ai" __license__ = "Apache-2.0"