New TF Lora Testing Results (NOT FINAL)
Added 2023-10-27 06:39:47 +0000 UTCThis isn't the final version of the lora model. I trained two separate models containing different parts of the overall data set, this way it's a lot quicker for me to train the lora, try it out, see what works what doesn't then re-train it, rather than re-training the AI on the entire data set each time. This should save me a lot of time :)
While I was going to delete some which didn't come out so great, I decided to leave everything in, this way people can see whats up and what I've been working on.
Some of the things that need doing.
-Re-training at a higher resolution
-More close up profile like tfs in the data set
-Few adjustments within captions for data set
Might not be fully perfect yet, but does make me excited for what the final product in the end will be like!!
(Side note: I did want the fox TF set to be first instead of the renamon set, but it seemed to upload in its own order lol, and yes these are based on old scenes/prompts I have generated in the past, so they might look a little similar, I did this for reference)
