Download
1 variant available
5110 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.5K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
2460 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1 / 10 images
(77)
Jul 13, 2024

1.3K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
4.9K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
85.8K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K

License:
CreativeML Open RAIL++-Mモデルを手直した感触だが、画像が多いとprodigyの結果は良くない。枚数が多いなら、Lionが結果が良い。
今回の例:
--optimizer_type="Lion8bit" --lr_warmup_steps=100 --max_grad_norm=1 --text_encoder_lr=0.00015 --unet_lr=0.00015
I have a feeling that I reworked the model, but if there are a lot of images, prodigy does not give good results. If there are many images, Lion has better results.
Example of this time:
--optimizer_type="Lion8bit" --lr_warmup_steps=100 --max_grad_norm=1 --text_encoder_lr=0.00015 --unet_lr=0.00015

