-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Contact Details
_No response_
### What should this feature add?
Support SDXL Checkpoint VAEs. For example, t…
-
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…
-
I am trying to reimplement your code in PyTorch and I need to know what is the difference between your loss function and the loss function regarding the vanilla VAE? Based on my experience, your KL-di…
-
Thank you for your working and sharing. I learned a lot from them.
However, I have a question about the VAE implementation.
In VAE.py, you calculate the marginal_likelihood as an cross entropy:
…
-
Hi there, in train.py line 327, it seems like the 'logits' is directly used as the latent variable z. Based on the loss function KL_gaussian_sem it seems like that would be the log of mean of z.
Th…
-
Thank you the nice tutorial and supporting code. I made a plot (attached) of KL Loss vs iterations of your implementation and that of Keras ([blog](https://blog.keras.io/building-autoencoders-in-keras…
-
I've been trying to train diffusion model with stable audio 1.0 config, I also trained the autoencoder with the Stable Audio 1.0 VAE for 50k steps [autoencoder result](https://storage.googleapis.com/w…
-
What a great job. When i use A800 80G with default parameters to infer the 768P video, I find that the GPU memory increases first and then decreases. The step inference is fine, but an OOM error is re…
-
On Macos 13.1
I get this error when running this command
`
python -m python_coreml_stable_diffusion.torch2coreml --model-version stabilityai/stable-diffusion-2-base --convert-unet --convert-text…
-
Hi, LOVE this extension/implementation. It is currently the BEST solution for upscaling.
when setting the tile size, sometimes when upscaling, I can get into 100s of tiles, could there be a calcula…