Textual Inversion embedding trained on 28 images of Canadian singer Avril Lavigne, from various professional photoshoots and stage appearances.
Tested and working well in RinaMix2, simpmaker3k1, DantionRealMix, deliberate, Realistic Vision 1.3, Epic Diffusion. May work well in others, but you may need to fiddle with the weight in them.
Description
base version. 3K steps. 28 images, batch size 4, 428 epochs.
FAQ
Comments (4)
Can you say more about the learning rate you apply to train this model please ?
how to use it?
I think this is a Lora, not an embedding, embedding you would put into the negative prompt but it does not work there. I put it in the positive prompt and it works. Embeddings go into the negative prompt, that is why I think this is a Lora.
Can you do Dasha Taran? She's known as one of the most beautiful faces in the world. Can get high quality from her videos, or instagram. I love the work you do!!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.