7th attempt. Decent enough to release, but not very flexible. You'll have to play with the weights depending on the rest of your prompt. I find training on old photos gives flatter results, to tried setting <code>noise_offset</code> to 0.1. I think it gives more pop to the images, but sometimes they get too dark. It overtrained on smiling pics so you may need to place <code>smiling</code> in negative prompt.

Romy Schneider

PICS: 86BLIP: 10 - 75<code>num_repeats</code>: 15<code>num_epochs</code>: 8<code>min_snr_gamma</code>: 5<code>noise_offset</code>: 0.1<code>network_dim</code>: 64<code>network_alpha</code>: 16steps: 12% 217/1736 [04:53&lt;34:15, 1.35s/it, loss=0.105]steps: 25% 434/1736 [09:59&lt;29:57, 1.38s/it, loss=0.101]steps: 38% 651/1736 [15:03&lt;25:06, 1.39s/it, loss=0.103]steps: 50% 868/1736 [20:08&lt;20:08, 1.39s/it, loss=0.1]steps: 62% 1085/1736 [25:13&lt;15:08, 1.39s/it, loss=0.105]steps: 75% 1302/1736 [30:18&lt;10:06, 1.40s/it, loss=0.0964]steps: 88% 1519/1736 [35:23&lt;05:03, 1.40s/it, loss=0.0991]steps: 100% 1736/1736 [40:27&lt;00:00, 1.40s/it, loss=0.0949]

Files with SHA256: 0d36c943384be1d78edffc017e76dc78e28d038099cad37fe2f28020d32ebba9