LibMTL.weighting.EW

class EW[source]

Bases: LibMTL.weighting.abstract_weighting.AbsWeighting

Equal Weighting (EW).

The loss weight for each task is always 1 / T in every iteration, where T denotes the number of tasks.

backward(losses, **kwargs)[source]
Parameters:

  • losses (list) – A list of losses of each task.

  • kwargs (dict) – A dictionary of hyperparameters of weighting methods.