torchmetrics

torchmetrics copied to clipboard

Investigate MPS support

3

## 🚀 Feature We need to investigate support for the new 'mps' backend. I assume this will not be working, as many operations are not yet supported. However, we should...

justusschock

Typo in the docstring in precision_recall_curve.py

1

There's a [typo](https://github.com/Lightning-AI/metrics/blob/58e7ac124ca8800c507558c5b0b477e5fa38a2af/src/torchmetrics/functional/classification/precision_recall_curve.py#L29) in the URL mentioned in the docstring. Thanks!

saikatkumardey

Added generalized dice score metric

20

## What does this PR do? Fixes #1089 . Didn't manage to make it work for multidimensional multilabel scenarios. However, that shouldn't a very common use case. ## Before submitting...

jlcsilva

mIoU

2

Where is the semantic segmentation mIou?

Lee-Akita

JaccardIndex then multilabel=True not working

1

## 🐛 Bug JaccardIndex is not correctly coded when multilabel=True and average is different than "none". ### To Reproduce Instantiate JaccardIndex(..., multilabel=True, average="micro") and call it as usually with a...

claverru

Links from documentation seem broken

2

When looking at specfic metric in documentation, some links to github source code seem to be broken. For example, On: https://torchmetrics.readthedocs.io/en/stable/regression/mean_squared_error.html When I click on **source** for `MeanSquaredError`, it leads...

masadcv

MAP: change bool to float32

3

## What does this PR do? Fixes #1147 ## Before submitting - [x] Was this **discussed/approved** via a Github issue? (no need for typos and docs improvements) - [x] Did...

dreaquil

Fleiss Kappa

9

## 🚀 Feature Fleiss Kappa [Wikipedia](https://en.wikipedia.org/wiki/Fleiss%27_kappa) ### Motivation Fleiss Kappa is a metric of inter-rater agreement between $k$ raters. It is useful in many areas for example: combining multiple measures...

cemde

Discrepancy between values with metrics and detectron2 cocoeval

1

## 🐛 Bug Training a model with detectron2 and evaluating it with the built-in cocoeval results in different mAP estimates for both `box` and `segm` metric cases. ### To Reproduce...

23pointsNorth

RuntimeError when using MAP-metric

17

## 🐛 Bug Hi! I am training a detection model and use MAP-metric during validation. I got the following error at the validation_step: `RuntimeError: expected scalar type Float but found...

V-Soboleva