Detailed Full Workflow

• Medium article : https://medium.com/@furkangozukara/ultimate-flux-lora-training-tutorial-windows-and-cloud-deployment-abb72f21cbf8

• Windows main tutorial : https://youtu.be/nySGu12Y05k

• Cloud tutorial for GPU poor or scaling : https://youtu.be/-uhL2nW7Ddw

• Full detailed results and conclusions : https://www.patreon.com/posts/111891669

• Full config files and details to train : https://www.patreon.com/posts/110879657

• SUPIR Upscaling (default settings are now perfect) : https://youtu.be/OYxVEvDf284

Details

• I used my Poco X6 Camera phone and solo taken images

• My dataset is far from being ready, thus I have used so many repeating and almost same images, but this was rather experimental

• Hopefully I will continue taking more shots and improve dataset and reduce size in future

• I trained Clip-L and T5-XXL Text Encoders as well

• In the above shared images the 19th image is the used dataset, 256 images, and 20th image is the comparison with 15 images training dataset and several checkpoints of newest training

• Since there was too much push from community that my workflow won’t work with expressions, I had to take a break from research and use whatever I have

• I used my own researched workflow for training with Kohya GUI and also my own self developed SUPIR app batch upscaling with face upscaling and auto LLaVA captioning improvement

• Download images to see them in full size, the last provided grid is 50% downscaled

Workflow

• Gather a dataset that has expressions and perspectives that you like after training, this is crucial, whatever you add, it can generate perfect

• Follow one of the LoRA training tutorials / guides

• After training your LoRA, use your favorite UI to generate images

• I prefer SwarmUI and here used prompts (you can add specific expressions to prompts) including face inpainting : https://gist.github.com/FurkanGozukara/ce72861e52806c5ea4e8b9c7f4409672

• After generating images, use SUPIR to upscale 2x with maximum resemblance

Short Conclusions

• Using 256 images certainly caused more overfitting than necessary

• I had to make prompts more detailed about background / environment to reduce impact of overfit, used Claude 3.5 (like ChatGPT)

• Still FLUX handled this massive overfit dataset excellently

• It learnt my body shape perfectly as well (muscular + some extra fat)

• It even learnt even my broken teeth or my forehead veins perfectly

• The outputs are much more lively and realistic and has better anatomy

• I couldn’t get such quality photo in a professional studio as in image 18 — the quality and details next level

• Since dataset was collected at different days, weeks, months, the hair, the weight of me, the skin color was not consistent, which caused some different hair style and length or skin color at inference :D