Open chr5tphr opened 11 months ago
With the implementation of #185 allowing for gradient computation wrt. the parameters given an attribution with canonizers, the BatchNorms seem to be leaking attribution even with zero_params='bias' on ResNet18.
zero_params='bias'
With the implementation of #185 allowing for gradient computation wrt. the parameters given an attribution with canonizers, the BatchNorms seem to be leaking attribution even with
zero_params='bias'
on ResNet18.