<h1 id="heading-861">Intro</h1><p>Brigitte Bardot is a French animal rights activist and former actress, singer, and model. Famous for portraying sexually emancipated characters, often with hedonistic lifestyles, she was one of the best known activists in the sexual revolution of the 1950s–1970s. Although she withdrew from the entertainment industry in 1973, she remains a major popular culture icon and a noted figure in ushering in the sexual revolution. She has acted in 47 films, performed in several musicals, and recorded more than 60 songs.</p><h1 id="heading-862">Pre Training</h1><p>I gathered 25 images of Brigitte Bardot. I only trained on HD images with color. I used Blip captioning to generate the filewords and edited each individually to reduce potential hallucinations. I added a 1 to the end of Bardot on the embedding name to ensure there is no mixing with previously trained images of her.</p><h1 id="heading-863">Training</h1><p>I used 0.005:100:0,0.0025:250,0.001:500,0.0005:1000,0.00025 for my learning rate. I am going for 2.5K training steps total. I am using a batch size of 1 with Gradient Accumulation Steps set to 3. I am running on a RTX 4090 on the cloud. I am using 12.5 out of 24 GB. The estimated time of completion is .5 hours. For the embedding I am using 5 vectors per token. I switched to SD 1.5 EMA Only model for training.</p><p></p><h2 id="heading-864">If you like this model and are interested in generating more visit me <a target="_blank" rel="ugc" href="http://ianolmstead-com.vercel.app">on my website</a>!</h2><p><br /></p>

Files with SHA256: d625c7eb15c5ffa53789f0e5130f268cea178ddf6bd81eb855dc1437f9735525