Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
mfai documentation
Logo
mfai documentation
  • Installation
  • Quickstart
  • Use Cases
  • NamedTensor
  • Models
    • ModelABC
  • Losses
    • GridCellLoss
    • BaseNet
    • LPIPS
    • LinLayers
    • PerceptualLoss
    • VGG16
    • DiceLoss
    • SoftBCEWithLogitsLoss
    • SoftCrossEntropyLoss
  • Transforms
    • DimensionSubSampler
    • MeanDimensionSubsampler
    • RandomCropWithMinPositivePixels
  • Metrics
    • CSINeighborhood
    • FAR
    • FNR
    • FSS
    • PR_AUC
  • Lightning
    • CLIPLightningModule
    • DGMRLightningModule
    • SegmentationLightningModule
    • MLFlowSaveConfigCallback
    • MLFlowSystemMonitorCallback
    • SaveCLIPVisualEncoderWeights
  • Reference
    • mfai
      • encoding
      • http
      • pytorch
        • callbacks
        • dummy_dataset
        • lightning_modules
          • clip
          • gan_dgmr
          • segmentation
        • losses
          • dgmr
          • perceptual
          • toolbelt
        • lr_scheduler
        • metrics
        • namedtensor
        • padding
        • transforms
      • tensorflow
  • About
  • Acknowledgements
  • Contributing
Back to top
View this page

losses¶

dgmr

Module for various loss functions used with DGMR GAN.

perceptual

toolbelt

'Classical' losses imported from https://github.com/BloodAxe/pytorch-toolbelt and adapted to our projects.
Next
dgmr
Previous
segmentation
Copyright © 2026, Météo-France AI Lab
Made with Sphinx and @pradyunsg's Furo
Météo-France