This was a test to see if tagging certain images as poses makes it available to generation... like a weaker version of training in one image only? IDK

Every image in the training data is . If you seek creative loras, you will find none here. Not sure it worked, as you can probably see from the samples, consider them as placeholders ( placeholders lmao)

Did a bit of testing, results are not what I intended. Pictures look good, but it's just not what I expected. None of the pictures in the showcase matches the pose tag, but they look good for some reason. Maybe it's PDXL magic and the score_x tags, because attempting to prompt without the tags generates mangled nightmare fuel