Official Tensorflow implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation (ICLR 2020)
MIT License
6.17k
stars
1.05k
forks
source link
Is there a general principle for adjusting the weight #102
When and how should I adjust the weights for different terms of losses according to our specific dataset, like a small dataset, or a dataset with large geometric shape change?
Maybe it's because of the small number of data. Or reduce the
cam_weight.Originally posted by @taki0112 in https://github.com/taki0112/UGATIT/issues/41#issuecomment-521521473
When and how should I adjust the weights for different terms of losses according to our specific dataset, like a small dataset, or a dataset with large geometric shape change?