Method of making this Lora:<ol><li>Downloaded ~500 of top results for <a target="_blank" rel="ugc" href="https://www.pexels.com/search/portrait/">https://www.pexels.com/search/portrait/</a></li><li>Resized them to max 1024px tall/width (I used powertoys bulk resize)</li><li>Manually went through them and discarded photos which were too abstract, had multiple people in it, or the faces were not fully visible</li><li>Used Kohya_ss to caption the photos with BLIP</li><li>Trained here on CivitAI (SDXL default settings)</li></ol>Intended to be used with face swap / custom face Lora.Enjoy.

Files with SHA256: c5e0e7c9d2048787929ba2fb1baf24dc4b2896c7d05a6a0871cc86d4e6881dbe