Native support for Torchmetrics MetricCollection

[SkafteNicki]

🚀 Feature

Add native support in self.log for the MetricCollection object from Torchmetrics: https://torchmetrics.readthedocs.io/en/latest/pages/overview.html#metriccollection

Motivation

The MetricCollection object is used to chain together multiple metrics into a single object that can be update at once:

import torchmetrics as tm
metrics = tm.MetricCollection([tm.Accuracy(), tm.Precision(), tm.Recall()])
metrics.update(preds, target)

however, when one want to log such an object in PL it is often done on the individual metric:

self.log_dict({k:v for k,v in collection.items()})

However, this actually does not work with the compute_group feature that we implemented in v0.8 for faster computation of metrics that share the same computations.

This issue proposed that PL natively supports MetricCollection in the same way that Metric is support in self.log.

Pitch

Alternatives

Additional context

Already discussed on slack with @Borda and @justusschock

---

If you enjoy Lightning, check out our other projects! ⚡

• Metrics: Machine learning metrics for distributed, scalable PyTorch applications.

• Lite: enables pure PyTorch users to scale their existing code on any kind of device while retaining full control over their own loops and optimization logic.

• Flash: The fastest way to get a Lightning baseline! A collection of tasks for fast prototyping, baselining, fine-tuning, and solving problems with deep learning.

• Bolts: Pretrained SOTA Deep Learning models, callbacks, and more for research and production with PyTorch Lightning and PyTorch.

• Lightning Transformers: Flexible interface for high-performance research using SOTA Transformers leveraging Pytorch Lightning, Transformers, and Hydra.

cc @borda @carmocca @edward-io @ananthsub @rohitgr7 @kamil-kaczmarek @Raalsky @Blaizzy @akihironitta

[stale[bot]]

This issue has been automatically marked as stale because it hasn't had any recent activity. This issue will be closed in 7 days if no further activity occurs. Thank you for your contributions, Pytorch Lightning Team!

[carmocca]

Setting this to "future" for now as it might not make the 1.7 release