Priming an image/Setting optimization goal issues

lucidrains / deep-daze

Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network). Technique was originally created by https://twitter.com/advadnoun

MIT License

4.37k stars 326 forks source link

Priming an image/Setting optimization goal issues #127

Open MightyObserver opened 3 years ago

MightyObserver commented 3 years ago

Im a beginner so sorry if this is some cringe inducing question, but I genuinely cannot get the priming and optimization parameters to go through, Issue I attached a photo of the error message, I think I'm misunderstanding the command to prime or optimize the AI, any help is appreciated and Thank you in advance.

NotNANtoN commented 3 years ago

If you want to optimize based on an image (Seb.jpg) you need to tell DeepDaze to use it as an image, otherwise it will interpret your path as a text string.

It is easiest if you just copy the image into the repository folder and then you can run: imagine --img=Seb.jpg or imagine YOUR_TEXT --img=Seb.jpg or imagine YOUR_TEXT --start_image_path=Seb.jpg