featuregood first issue
Metriche repository
- Star
- (15.550 stelle)
- Metriche merge PR
- (Merge medio 2g 22h) (129 PR mergiate in 30 g)
Descrizione
Feature description
Add metrics for assessing image quality, including deep learning metrics.
- L1/MAE loss, torch, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4341
- L2/MSE loss, torch, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4341
- PSNR, Wikipedia, scikit-image, PyTorch-ignite, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4379
- SSIM, Wikipedia, scikit-image, PyTorch-ignite, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4396
- MS-SSIM, lightning.ai, pytorch-msssim
- Closed via https://github.com/tracel-ai/burn/pull/4555
- Smooth L1, torch
- Closed via https://github.com/tracel-ai/burn/pull/4547
- Gram Matrix Loss, torch tutorial, piq
- Closed via https://github.com/tracel-ai/burn/pull/4595
- FID, lightning.ai, pytorch-fid, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4644
- LPIPS, lightning.ai, torch-image-metrics
- Closed via https://github.com/tracel-ai/burn/pull/4403
- DISTS, lightning.ai
- Closed via https://github.com/tracel-ai/burn/pull/4574
- A-FINE, IQA-PyTorch
Feature motivation
When training visual AI models the loss function may include various image metrics. Users would have to re-implement these image metrics unless readily available.
Examples
Take for example the impressive recent ml-sharp which uses at least PSNR, SSIM, LPIPS, DISTS.