v5.0a
"unetLR": 0.0005,
"clipSkip": 2,
"networkDim": 128,
"numRepeats": 10,
"resolution": 1024,
"lrScheduler": "cosine_with_restarts",
"noiseOffset": 0.1,
"networkAlpha": 64,
"optimizerType": "AdamW8Bit",
"textEncoderLR": 0.00005,
bigger dataset v3.0b:Text Encoder learning rate:0.000005Unet learning rate:0.0001LR Scheduler:cosineOptimizer:AdamW8bitNetwork Dim:64Network Alpha:32v3.0a:Bigger training dataset0.8 have best result1 look weird for meText Encoder learning rate:0.00001Unet learning rate:0.0001LR Scheduler:constantOptimizer:AdamW8bitNetwork Dim:32Network Alpha:16
