How to use autoencoder_fix_kl-f8-trinart_characters.ckpt?
This is a question about the following paragraph in the Readme file:
'''
Custom autoencoder
Note: The autoencoder uploaded here is the same checkpoint as v1.
We also provide a separate checkpoint for the custom KL autoencoder. As suggested by the Latent Diffusion paper, we found that training the autoencoder and the latent diffusion model separately improves the result. Since the official stable diffusion script does not support loading the other VAE, in order to run it in your script, you'll need to override state_dict for first_stage_model. The popular WebUI has the script to load separate first_stage_model parameters.
'''
Does this mean that, in order to use the derrida_final Checkpoint, we also need to load the autoencoder_fix_kl-f8-trinart_characters.ckpt via some script in the WebUI that loads separate first_stage_model parameters? If so, how is this done? From my looking at the popular WebUI, I couldn't find any settings to change the first_stage_model parameters.
@07mk you just need to put the autoencoder in the vae folder under webui's models/vae folder and then select it in the settings of the webui.
Alternatively, name it the same thing as the model, e.g. derrida_final and add the ".vae.pt" to the end, replace .ckpt, so it would look like "derrida_final.vae.pt" in the model folder alongside the checkpoint of the same name. The WebUI will load a vae that's alongside a model that has the same name as the model.