Weighted learning of TIs and HNs

fix arg for train_hypernetwork api

Fix prompt matrix #rows/#cols when using hires

Add Image CFG Scale to XYZ Grid

Fix: show correct help message on --help

Update ui_extra_networks.py to fix div id's that have spaces in them

Update batch count/size hints

Calls modules.sd_vae.refresh_vae_list(), fixes VAE list not updating

Add missing imports

Convert image from RGBA to RGB before saving

Fix hypernetworks and instruct pix2pix not working with `--upcast-sampling`

Add cfg_denoised_callback

Revert "Aspect ratio sliders"

Aspect ratio sliders

Convert 16-bit greyscale to 8-bit when saving as JPEG

Fix face restorers setting

reword settings for 4chan export, remove unneded try/excepts, add try/except for actually saving JPG

Configurable JPG downscale threshold

add version to extensions table

Fix DPM++ SDE not deterministic across different batch sizes (#5210)

Fix params.txt saving for infotexts modified by process_batch

Fixed missing part in the Linux installation instructions

Download model if none are found

Apply extra networks per-batch instead of per-session (fixes wildcards)

store and print real torch version