If the prompt is: a man wearing {red|blue|yellow|green} shirt, and I start a batch count of 50, batch size of 1, then the 50 generated image will all be one of the four colors, e.g. a man wearing red shirt 50 times.
What I would expect it to do instead is having all four colors randomized for the 50 count.
Is there any way to get around this limitation?
If the prompt is:
a man wearing {red|blue|yellow|green} shirt, and I start a batch count of 50, batch size of 1, then the 50 generated image will all be one of the four colors, e.g.a man wearing red shirt50 times.What I would expect it to do instead is having all four colors randomized for the 50 count.
Is there any way to get around this limitation?