Hello everyone. I am Dr. Furkan Gözükara. PhD Computer Engineer. SECourses is a dedicated YouTube channel for the following topics : Tech, AI, News, Science, Robotics, Singularity, ComfyUI, SwarmUI, ML, Artificial Intelligence, Humanoid Robots, Wan 2.2, FLUX, Krea, Qwen Image, VLMs, Stable Diffusion
the best would be Deepface Lab. if you got tons of photos of a person and willing to spend few days in training. Secon best is FaceFusin/roop/rope/reactor takes 1 photo and lightning fast but low quality 128x128 only. DeepfaceLab works up to 512x512 for the face
Is there a bug or something with Kohya? I start the training with 24GB Tier3, everything is loaded, memory looks fine, gpu utilisation is fine, yet I'm getting horrible speeds