Published January 15, 2022
| Version v0.7.0rc1
Software
Open
PyTorchLightning/metrics: Feature teaser
Creators
- Jirka Borovec1
- Nicki Skafte Detlefsen
- Daniel Stancl2
- Maxim Grechkin
- Luca Di Liello
- Changsheng Quan3
- Justus Schock4
- William Falcon5
- Teddy Koker6
- Tobias Kupek7
- edenlightning
- thomas chaton8
- Aki Nitta9
- Tadej Svetina10
- Adrian Wälchli5
- Jeff Yang
- Rohit Gupta11
- karthikrangasai12
- Ananya Harsh Jha
- Dusan Drevicky
- Ethan Harris13
- Maheshwari Natarajan14
- twsl
- Ashutosh Kumar
- Bhadresh Savani15
- Cookie_thief
- Pranjal Gulati16
- Santiago Castro17
- 1. Grid.ai | PytorchLightning
- 2. Seznam.cz | Charles University
- 3. WestlakeU
- 4. @rwth-lfb , @PyTorchLightning & University Hospital Düsseldorf
- 5. Grid AI
- 6. @mit-ll
- 7. @hal9000-swarm
- 8. Grid.ai | Pytorch Lightning
- 9. @PyTorchLightning
- 10. @jina-ai
- 11. @gridai @pytorchlightning
- 12. BITS Pilani Hyderabad Campus
- 13. Grid.ai
- 14. Zoho
- 15. @Quanitphi
- 16. @dsgiitr
- 17. @MichiganNLP
Description
[0.7.0] - 2022-01-DD Added
- Added NLP metrics:
- Added
MultiScaleSSIM
into image metrics (#679) - Added Signal to Distortion Ratio (
SDR
) to audio package (#565) - Added
MinMaxMetric
to wrappers (#556) - Added
ignore_index
to retrieval metrics (#676) - Added support for multi references in
ROUGEScore
(#680) - Added a default VSCode devcontainer configuration (#621)
- Scalar metrics will now consistently have additional dimensions squeezed (#622)
- Metrics having third party dependencies removed from global import (#463)
- Untokenized for
BLEUScore
input stay consistent with all the other text metrics (#640) - Arguments reordered for
TER
,BLEUScore
,SacreBLEUScore
,CHRFScore
now expect input order as predictions first and target second (#696) - Changed dtype of metric state from
torch.float
totorch.long
inConfusionMatrix
to accommodate larger values (#708) - Unify
preds
,target
input argument's naming across all text metrics (#723, #727)bert
,bleu
,chrf
,sacre_bleu
,wip
,wil
,cer
,ter
,wer
,mer
,rouge
,squad
- Renamed IoU -> Jaccard Index (#662)
- Renamed
WER
->WordErrorRate
andwer
->word_error_rate
(#714) - Renamed correlation coefficient classes: (#710)
MatthewsCorrcoef
->MatthewsCorrCoef
PearsonCorrcoef
->PearsonCorrCoef
SpearmanCorrcoef
->SpearmanCorrCoef
- Renamed audio STOI metric: (#753, #758)
audio.STOI
toaudio.ShortTermObjectiveIntelligibility
functional.audio.stoi
tofunctional.audio.short_term_objective_intelligibility
- Renamed audio PESQ metrics: (#751)
functional.audio.pesq
->functional.audio.perceptual_evaluation_speech_quality
audio.PESQ
->audio.PerceptualEvaluationSpeechQuality
- Renamed audio SDR metrics: (#711)
functional.sdr
->functional.signal_distortion_ratio
functional.si_sdr
->functional.scale_invariant_signal_distortion_ratio
SDR
->SignalDistortionRatio
SI_SDR
->ScaleInvariantSignalDistortionRatio
- Renamed audio SNR metrics: (#712)
functional.snr
->functional.signal_distortion_ratio
functional.si_snr
->functional.scale_invariant_signal_noise_ratio
SNR
->SignalNoiseRatio
SI_SNR
->ScaleInvariantSignalNoiseRatio
- Renamed F-score metrics: (#731, #740)
torchmetrics.functional.f1
->torchmetrics.functional.f1_score
torchmetrics.F1
->torchmetrics.F1Score
torchmetrics.functional.fbeta
->torchmetrics.functional.fbeta_score
torchmetrics.FBeta
->torchmetrics.FBetaScore
- Renamed Hinge metric: (#734)
torchmetrics.functional.hinge
->torchmetrics.functional.hinge_loss
torchmetrics.Hinge
->torchmetrics.HingeLoss
- Renamed image PSNR metrics (#732)
functional.psnr
->functional.peak_signal_noise_ratio
PSNR
->PeakSignalNoiseRatio
- Renamed image PIT metric: (#737)
torchmetrics.functional.pit
->torchmetrics.functional.permutation_invariant_training
torchmetrics.PIT
->torchmetrics.PermutationInvariantTraining
- Renamed image SSIM metric: (#747)
torchmetrics.functional.ssim
->torchmetrics.functional.scale_invariant_signal_noise_ratio
torchmetrics.SSIM
->torchmetrics.StructuralSimilarityIndexMeasure
- Renamed detection
MAP
toMeanAveragePrecision
metric (#754) - Renamed Fidelity & LPIPS image metric: (#752)
torchmetrics.image.FID
->torchmetrics.image.FrechetInceptionDistance
torchmetrics.image.KID
->torchmetrics.image.KernelInceptionDistance
torchmetrics.image.LPIPS
->torchmetrics.image.LearnedPerceptualImagePatchSimilarity
- Removed
embedding_similarity
metric (#638) - Removed argument
concatenate_texts
fromwer
metric (#638) - Removed arguments
newline_sep
anddecimal_places
fromrouge
metric (#638)
- Fixed MetricCollection kwargs filtering when no
kwargs
are present in update signature (#707)
Files
PyTorchLightning/metrics-v0.7.0rc1.zip
Files
(855.0 kB)
Name | Size | Download all |
---|---|---|
md5:4535f512e3952f8f7f33d08cc212e0e1
|
855.0 kB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/PyTorchLightning/metrics/tree/v0.7.0rc1 (URL)