Support basic plotting for some metrics

[SkafteNicki]

🚀 Feature

Add plotting functionality for metrics where it makes sense.

Motivation

Some metrics like confusion matrix, roc, pr-curve are non-scalar metrics that are best inspected when plotted. It would be great if we could support some basic functionality for metrics where it makes sense to plot something (there are probably not that many).

Pitch

Either have a utility module:

from torchmetrics.utils.plotting import plot_confusion_matrix
metric = ConfusionMatrix()
output = metric(preds, target)
plot_confusion_matrix(output)

or incorporate directly into each modular metric

metric = ConfusionMatrix()
output = metric(preds, target)
metric.plot(output)

Alternatives

Keep as it is.

Additional context

[github-actions[bot]]

Hi! thanks for your contribution!, great first issue!

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[andredantasrocha]

hi guys, any updates on this? it would be amazing to have a feature like that :)

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.