Prompt Trigger:
Base Character:
Slo1h_M1n1, chibi, bear girl, bear ears, navel, animal ears, fur collar, white skin, animal nose, two-tone skin, neck fur, koala, thick eyebrows, purple skin, blue button, Eyes (Optional):
purple eyes, half-closed eyes,Negative Prompt:
furry female, body fur,Description
### Training Information (Anima DiT)
+ Output Name: bf16_anima_V1_sloth_mini
+ Network Module: networks.lora_anima
+ Epochs: 8
+ Max Train Steps: Auto (Epoch)
+ Learning Rate: 0.0001
+ Text Encoder LR: 5e-05
+ Optimizer: AdamW8bit
+ Mixed Precision: bf16
+ LR Scheduler: constant
+ LR Warmup Steps: 0
+ Rank: 16/16 (Dim, Alpha)
### Anima Arguments:
+ timestep_sample_method: logit_normal
+ discrete_flow_shift: 3
+ weighting_scheme: logit_normal
### Dataset Info (dataset.toml):
+ Batch Size: 1
+ Resolution: 1024x1024
+ Steps (calculated): 512
+ Bucketing: True
### Datasets Parameter:
+ Mini_from : images: 13 | repeats: 4 | keep_tokens: 2 | flip_aug: False | shuffle_caption: True
+ Mini_form_Other : images: 9 | repeats: 1 | keep_tokens: 1 | flip_aug: True | shuffle_caption: True
+ Mini_from_AI : images: 3 | repeats: 1 | keep_tokens: 1 | flip_aug: False | shuffle_caption: False