where was cur_weights from?

chufengt / ALM-pedestrian-attribute

Code for the paper "Improving Pedestrian Attribute Recognition With Weakly-Supervised Multi-Scale Attribute-Specific Localization", ICCV 2019, http://arxiv.org/abs/1910.04562.

Apache License 2.0

188 stars 66 forks source link

where was cur_weights from? #49

Closed eeric closed 3 years ago

eeric commented 3 years ago

cur_weights = torch.exp(target + (1 - target 2) self.weights)

chufengt commented 3 years ago

cur_weights are used to control the sample imbalance. See equation (5) in the paper for how to calculate the weights.