Sign In

Memoji

53
910
8
Updated: Nov 3, 2023
style3demojicartoonmemoji
Verified:
SafeTensor
Type
LoRA
Stats
910
Reviews
Published
Oct 24, 2023
Base Model
SDXL 1.0
Training
Steps: 8,120
Epochs: 5
Trigger Words
sico style
Hash
AutoV2
EA8C050A6D
default creator card background decoration
PixpiAI's Avatar
PixpiAI

Recommended weight: 0.4-0.8

Train on 1624 memoji images. Train logs:

running training / 学習開始
  num train images * repeats / 学習画像の数×繰り返し回数: 1624
  num reg images / 正則化画像の数: 0
  num batches per epoch / 1epochのバッチ数: 1624
  num epochs / epoch数: 5
  batch size per device / バッチサイズ: 1
  gradient accumulation steps / 勾配を合計するステップ数 = 1
  total optimization steps / 学習ステップ数: 8120
steps: 100%|██████████| 8120/8120 [3:10:10<00:00,  1.41s/it, Average key norm=0.726, Keys Scaled=0, loss=0.0341]