Experimenting with training a character LoRA starting from one image. Inspired by https://civitai.com/articles/3021/one-image-is-all-you-need
{
"unetLR": 0.0005,
"clipSkip": 1,
"loraType": "lora",
"keepTokens": 1,
"networkDim": ??,
"numRepeats": 20,
"resolution": 1024,
"lrScheduler": "cosine",
"minSnrGamma": 5,
"noiseOffset": 0.1,
"targetSteps": 1400,
"enableBucket": true,
"networkAlpha": 32,
"optimizerType": "Prodigy",
"textEncoderLR": 0.00005,
"maxTrainEpochs": 20,
"shuffleCaption": true,
"trainBatchSize": 8,
"flipAugmentation": false,
"lrSchedulerNumCycles": 3
}
Description
Trained on images from v1.0. Theoretically more robust and flexible.
FAQ
Details
Downloads
2,104
Platform
SeaArt
Platform Status
Available
Created
8/8/2024
Updated
9/15/2024
Deleted
-
Trigger Words:
pink eyes, ahoge, purple hair, blue streaked hair, pink gradient hair, hair bow, low-twintails
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.