drugforge.ml.early_stopping
Class for handling early stopping in training.
Classes
|
Class for handling early stopping in training based on improvement over best loss. |
|
Class for handling early stopping in training based on whether loss is still changing. |
|
Class for stopping based on the relative increase of val loss at epoch t from lowest val loss up to epoch t. |
|
Class for handling early stopping in training based on whether loss is still changing, with patience. |
|
Class for stopping based on quotient of generalization loss and training progress. |
|
Class for handling early stopping in training based on whether loss has been below a certain threshold for some number of epochs. |