//=time() ?>
More denim hotpants Yuuka.
#NovelAI #幽香 #ホットパンツ https://t.co/H8zlYkuM7N
The miata embedding should be okay now, at 45k total steps.
https://t.co/3n45LipS0E
#textualinversion #machinelearning https://t.co/t0kcUjx8K2
More happy-looking Yuukas!
#幽香 #NovelAI #yandere https://t.co/t0kcUjxGzA
Miata8674 embedding, 33k steps.
I suppose these count as representative samples.
#幽香 #NovelAI https://t.co/u7zfXjzAeb
23k steps @ high CFG.
Either the updates to UI/VAE or the embedding itself don't like it when it's close to something like "perfect face" in the prompt, it really messes with the result. See the first two images. The second pair are better ones...
#幽香 #textualinversion #NovelAI https://t.co/AxNibT64u4
At 11k steps there's very minimal similarity to miata; conversely, it's very noticeable with realistic art: I didn't highlight the eyes in the prompt at all!
I'm now positive that Google is indeed progressively lowering Colab limits with usage.
#textualinversion #machinelearning https://t.co/FIeOpnIaAy