Trying to work out training settings with Illustrious models; Overcooked it, the first time around. Was testing out Illustrious training, and went max on parameters with the first one. The v1.1 has more 'conventional' parameters.V2.0 onwards are Flux models trained on an entirely different dataset and use 'c0mix2' trigger keyword.

c0mix

Should be a more suitable fit for illustrious. It is bigger than the previous versions, both in terms of training dataset (200+ images) and the number of steps. Training parameters are listed below - "engine": "kohya", "unetLR": 1, "clipSkip": 1, "loraType": "lora", "keepTokens": 0, "networkDim": 64, "numRepeats": 3, "resolution": 1024, "lrScheduler": "cosine", "minSnrGamma": 5, "noiseOffset": 0.1, "targetSteps": 3045, "enableBucket": true, "networkAlpha": 32, "optimizerType": "Prodigy", "textEncoderLR": 1, "maxTrainEpochs": 10, "shuffleCaption": false, "trainBatchSize": 2, "flipAugmentation": false, "lrSchedulerNumCycles": 3

c0mix_ep-10-_v1.2.safetensors

v1.2

v2.5

v2.4 - Milo Manara Style

v2.3

v2.2

v2.1

v2.0

v1.1

It is a test attempt with Illustrious training. Will improvise upon it after more testing.

c0mix Illustrations - Illust Ed

Should be a more suitable fit for illustrious. It is bigger than the previous versions, both in terms of training dataset (200+ images) and the number of steps.

Training parameters are listed below - 
"engine": "kohya",
  "unetLR": 1,
  "clipSkip": 1,
  "loraType": "lora",
  "keepTokens": 0,
  "networkDim": 64,
  "numRepeats": 3,
  "resolution": 1024,
  "lrScheduler": "cosine",
  "minSnrGamma": 5,
  "noiseOffset": 0.1,
  "targetSteps": 3045,
  "enableBucket": true,
  "networkAlpha": 32,
  "optimizerType": "Prodigy",
  "textEncoderLR": 1,
  "maxTrainEpochs": 10,
  "shuffleCaption": false,
  "trainBatchSize": 2,
  "flipAugmentation": false,
  "lrSchedulerNumCycles": 3

Files with SHA256: 26ba08ef93001a8922f44b01eb67487fc02c9ded099239f6b5905183cf33a53e