Instaling Inpaint2/12/2024 ![]() I'm at the docker compose -profile auto up -build part and cmd is throwing this error at me: I ended up choosing hyper-v (had to reinstall docker) Go the long route and install/update WSL to 2.0 (which is quite difficult for the technically challenged like me) I would have had to either enable Hyper-V in Window's Program&Features (its not on by default) More parameter tweaks can be found in this guide (external):įor first timers like me, i feel this guide is incomplete, i had to go to docker's settings->resources then file sharing to enable file sharing with the folder where I extracted the webui zip to get the download up part to workīut first things first (for those who don't have wsl installed or up-to-date).Keep an eye on your RAM usage, depending on the amount, it's possible you will run out of RAM, and Windows will probably get very sluggish since it needs to swap constantly. ![]() Also check that the width & height in the model parameters are close / equal to that of the source image, or leave it at the default: 512x512.When not in use, it's best to stop the container, since it keeps a 'hold' on a lot of RAM and (graphics) VRAM. You should be able to use 'Docker Desktop' to start and stop the container from now on.Whenever you see 'OOM (Out of Memory)' in the console output, it most of the time means that you do not have enough (graphics) VRAM to run it.When asking for help, don't forget to add a screenshots of the console output, your masked image (do not post private images) and the model parameters.Sometimes it's needed to add extra words to the (positive) prompt, i.e. In some cases you need to add 'fingers', since Stable Diffusion is quite bad at generating fingers. I've already added some, like hair, water, etc. When the model is generating weird or unwanted things, add them to the 'Negative Prompt'. But look at the "in progress" images that are generated during the process. difficult to find with clothes that are dark. And sometimes: you forgot to mask some (tiny) area of the clothes.If you only mask clothes, the generator doesn't know about any skin colors, so it's best you include parts of the unclothed body (e.g. Mask more, go outside of the edges of the clothes.You can change the size of the brush at the right of the image. Now it's time to mask / paint over all the areas you want to 'regenerate', so start painting over all clothing. Open your preferred image in the left area. You can use any photo manipulation software, or use Stable Diffusion models work best with images with a certain resolution, so it's best to crop your images to the smallest possible area. at the edges), so the model will learn how to "fix" inconsistencies at the edges. Change masked padding, pixels this will include fewer/more pixels (e.g.Use a higher batch count to generate more images in one setting.Denoising strength: the lower the closer to the original image, the higher the more "freedom" you give to the generator.Latent Nothing: will generate no initial noise.Latent Noise: will generate new noise based on the masked content.This can be useful in some cases to keep body shape, but will more easily take over the clothes Original: will use the masked area to make a guess to what the masked content should include.Fill: will fill in based on the prompt without looking at the masked area.EXPERIMENT: don't take this guide word-by-word.These are just recommendations / what works for me, experiment / test out yourself to see what works for you. Denoise Strength: 0.75 (default, the lower you set this, the more it will look like the original masked area)
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |