LibMTL.weighting.DWA

class DWA[source]

Bases: LibMTL.weighting.abstract_weighting.AbsWeighting

Dynamic Weight Average (DWA).

This method is proposed in End-To-End Multi-Task Learning With Attention (CVPR 2019) and implemented by modifying from the official PyTorch implementation.

Parameters

T (float, default=2.0) – The softmax temperature.

backward(self, losses, **kwargs)[source]