pytorch/examples

VAE reconstruction loss (BCE)

开放

#460 创建于 2018年12月1日

 (2 条评论) (3 个反应) (0 位负责人)Python (9,429 个派生)batch import
good first issue

仓库指标

星标
 (21,634 个星标)
PR 合并指标
 (PR 指标待抓取)

描述

The current implementation uses

F.binary_cross_entropy(recon_x, x.view(-1, 784), reduction='sum')

as the reconstruction loss. The image x has pixel values in [0,1]. This is not the same as Bernoulli log likelihood. The images would have to binarized.

In Ladder Variational Autoencoders by Sonderby et al, they binarize the images as a Bernoulli sample after each epoch.

贡献者指南