training

Modules for training models.

graphnet.training manages model training and experiment logging.

Submodules

• callbacks
  • PiecewiseLinearLR
  • ProgressBar
  • GraphnetEarlyStopping
• labels
  • Label
  • Direction
  • Track
• loss_functions
  • LossFunction
  • MAELoss
  • MSELoss
  • RMSELoss
  • LogCoshLoss
  • CrossEntropyLoss
  • BinaryCrossEntropyLoss
  • LogCMK
  • VonMisesFisherLoss
  • VonMisesFisher2DLoss
  • EuclideanDistanceLoss
  • VonMisesFisher3DLoss
  • EnsembleLoss
  • RMSEVonMisesFisher3DLoss
• utils
  • collate_fn()
  • collator_sequence_buckleting
  • make_dataloader()
  • make_train_validation_dataloader()
  • get_predictions()
  • save_results()
  • save_selection()
• weight_fitting
  • WeightFitter
  • Uniform
  • BjoernLow