Trained on 40 images from 1920. works good between 0,5 and 1
528895_training_data.zip
1920_photography_style.safetensors