Open JiajianLu opened 9 months ago
Hi, from this line https://github.com/amazon-science/tabsyn/blob/main/tabsyn/vae/model.py#L367, each data x is encoded to a fixed z since it's only using the mu encoder. Should it be a sampled version? In other latent diffusion implementation, it seems a sampled version is used. thanks
I didn't try using the sampled version. There might be slights gap between their empirical performance. Thanks for your question and I will try it later.
Hi, from this line https://github.com/amazon-science/tabsyn/blob/main/tabsyn/vae/model.py#L367, each data x is encoded to a fixed z since it's only using the mu encoder. Should it be a sampled version? In other latent diffusion implementation, it seems a sampled version is used. thanks