❌💲 Commercial use of this model is not allowed
I like this bird species, but AI simply can't generate it, the closest you can get is an ambiguous corvid. So I made this LoRA to see if I can make it work.
🏷 DATASET TAGS (doesn't apply to Flux and ZIT, those use natural captions)
Tagging is a mix of autotagging and manual.
The whole dataset is tagged with "jackdaw, western jackdaw, eurasian jackdaw, coloeus monedula, corvus monedula, corvid, bird, feral, animal, animal focus, realistic, no humans".
Some pictures focus on the bird's head, most of the others use "fullbody".
Most images in the dataset show solo birds (tagged "solo, 1 jackdaw, 1 bird"), a few images show two ("duo, 2 jackdaws, 2 birds").
🖼 DATASET
122 images which are photos only (doesn't apply to the Flux version). I don't own anything.
Images don't contain signatures/watermarks.
Resolution is all over the place, flying pictures are lower quality overall, though some images were large enough to be automatically sized down.
Images don't show any humans or other distinct creatures.
Only adult birds with normal colouration. No eggs, hatchlings or fledglings, no colouration anomalies.
⚠ ISSUES
The bird's colouration most likely will be lost in cartoony results.
BONUS: I tried SD3.5L, but it doesn't work and I don't know why.
I deleted the LoRA for SD3.5L since it couldn't produce the jackdaw. You don't need a LoRA for generic corvids. I thought I could make a proper LoRA if I used all the 122 images and maybe increased the number of Epochs. But seeing the sample images, I knew that nothing helped... I'm not putting these useless LoRAs on here, but if you can use SD3.5L locally, you can see for yourself. I took time to correct the captions on 14 pages of images so there wouldn't be any incorrect bird names, so it wouldn't be called "large" or its eyes "yellow", and it was all pointless... How come Flux works with about 30 images, but SD3.5 doesn't with 100+? And I wondered why SD3.5 LoRAs were so scarce. Ain't gonna bother with it again anytime soon.
Description
Same dataset as for all the previous SD LoRAs.
Trained against NoobAI-XL (Epsilon-pred 1.1-Version) via Civitai.
FAQ
Comments (4)
A part of me thinks I should change the Noob LoRA's category to Illustrious. Noob still doesn't have its own category, and methinks most Noob finetunes are set to Illustrious. What do you think?
Upd: if anyone can use SD3.5 locally, please test this LoRA, I squeezed out one image with bad eyes from the generator, and now that I try again with modified settings it's not working again.
...................................
I recall when I first did some SD3 gens on site, they were generally much faster than SDXL. Now when I need images quickly to see how my LoRA works and the base model is too much for my GTX card, it's taking forever.
I simply don't want to post sample images I got prior to downloading the LoRA because they're always different from what it actually does.
The SD3.5 LoRA got unpublished because the images I added after didn't actually add to the showcase, and because I deleted the placeholder it was technically empty. But when I added images to SDXL LoRAs, they would always fill the showcase until it was full? What is this bullshit? I added a couple of DDIM samples to the showcase to get it back.
Anyway, I may retrain this version with the larger dataset used for SDXL versions and SD1.5 and natural language captions in the near future, while possibly unpublishing the beta version since it fails the one thing I want it to do.
Updated the NoobAI LoRA's info to say NoobAI