Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁
Check out all the Art styles I’ve made: https://civarchive.com/user/CitronLegacy/models?tag=style
This is trained on almost 600 pictures of anime figures.
Description
Trained with the CivitAI trainer
Trained on: 585 images
unetLR: 0.0005
clipSkip: 1
loraType: "lora"
keepTokens: 0
networkDim: 32
numRepeats: 1
resolution: 1024
lrScheduler: "cosine"
minSnrGamma: 0
noiseOffset: 0.03
targetSteps: 1463
enableBucket: true
networkAlpha: 32
optimizerType: "Prodigy"
textEncoderLR: 0.00005
maxTrainEpochs: 10
shuffleCaption: true
trainBatchSize: 4
flipAugmentation: false
lrSchedulerNumCycles: 3
FAQ
Comments (5)
CoF Lora when? xD
GOOOD! <3
Unfortunately, v1 Pony negatively impacts background quality too much even at 0.3 strength. I suspect that the training data might only include figurines with simple backgrounds.
Yes the training data mostly contained simple backgrounds. If you shop for anime figures online most of the pictures have simple backgrounds.
@CitronLegacy Definitely something to look into rectifying if you decide to make a v2 of this LoRA.
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.