Playing with the text to image functionality of stable diffusion left me a little frustrated because the turn around time is fairly slow on my machine, around two minutes per image (about double what I originally estimated). And I would get sidelined playing with words trying to finetune the generation. I decided I need to do then more in bulk and wrote up a little program to generate the prompts using a "word bingo" approach. This evolved over a day or so the copy the files and rename them with the prompt that was used.
I've now generated about 1500 images, sadly the first 700 or so I don't have a good record of the prompts. Anyway here is a small album of 32 of my favourites:
I will upload the code for my script that creates the prompts and copes and renames the files after I clean it up a bit.