extern/easydiffusion
1
0
Fork 1
mirror of https://github.com/easydiffusion/easydiffusion.git synced 2025-06-20 09:57:49 +02:00
Code Issues Packages Projects Releases Wiki Activity

Updated UI Overview (markdown)

SpecificKnot 2023-02-14 14:35:39 +00:00
parent 81d4dfacb1
commit 9656e6efae
1 changed files with 90 additions and 47 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

137
UI-Overview.md

@ -1,62 +1,105 @@
# Prompt and Base Image # Prompt and Base Image
![image](https://user-images.githubusercontent.com/2499585/200661360-90c87055-b351-4946-801b-fd053eb84a45.png)
* Prompt: A textual prompt for stable diffusion to render - [[More on prompts|Writing prompts]] ![image](https://user-images.githubusercontent.com/110454200/213925096-dd1a38f4-9c4d-4c44-abbd-b16177f96c1f.png)
* Load from a file: Load textual prompt from a .txt-file - one prompt per line
* Negative Prompt: A textual prompt for stable diffusion to avoid - [[More on negative prompts|Writing-prompts#negative-prompts]] **Prompt**: A textual prompt. Enter something here for stable diffusion to render.
* Initial Image: A selector for a base image
**Load from a file**: Load a textual prompt from a .txt file - one prompt per line.
**Negative Prompt**: A textual prompt that stable diffusion will try to avoid when rendering your image.
**Initial Image**: Stable-Diffusion will use an existing `Initial Image` to generate your image. Choose an image from your computer.
# Image Settings # Image Settings
![image](https://user-images.githubusercontent.com/2499585/200583269-79c246a1-eadb-4685-bcb4-3ce3c67a9a95.png)
![image](https://user-images.githubusercontent.com/110454200/218768343-8a1b74f1-2bc4-4bab-afce-94d470d8cc34.png)
## Image Settings ## Image Settings
* Seed: A number to prime the random generator with **Seed**: A number that will be used to randomize your image.
* Number of images:
* (total) A number for how many images are to be generated in total **Number of Images**:
* (in parallel) A number for how many images are to be generated at the same time * (total): How many images are to be generated in total.
* Model: A drop-down-menu for selecting data models - [[More on data models|Custom-Models]] * (in parallel): How many images are to be generated at the same time.
* Custom VAE: A drop-down-menu for selecting Variational Auto Encoder - [[More on VAE|VAE-Variational-Auto-Encoder]]
* Image Size: drop-down-menus to select the resolution of the image to be generated **Model**: Select a data model which will be used to determine how to render your image.
* Inference steps: A number for how many steps stable diffusion should use - [Example](https://getimg.ai/guides/interactive-guide-to-stable-diffusion-steps-parameter)
* Guidance Scale: A slider & number field that controls how much the imgage generation should follow the prompt - [Example](https://getimg.ai/guides/interactive-guide-to-stable-diffusion-guidance-scale-parameter) **Custom VAE**: Selecting a `Variational Auto Encoder`.
* Output format: A drop-down-menu for selecting either jpeg or png
**Image Size**: Select what the resolution of the image should be.
**Inference Steps**: Stable Diffusion works by generating images in multiple attempts. The number of attempts is what's called `Inference Steps`.
[Example](https://getimg.ai/guides/interactive-guide-to-stable-diffusion-steps-parameter)
**Guidance Scale**: Controls how much the image generation should follow the prompt. [Example](https://getimg.ai/guides/interactive-guide-to-stable-diffusion-guidance-scale-parameter)
**Output Format**: To select the file format of the rendered image. Can be either `JPEG` or `PNG`.
## Render Settings ## Render Settings
* Show a live preview: Shows a live preview if enabled, otherwise will display the image when finished
* Fix incorrect faces and eyes: Uses GFPGAN - a face restauration algorithm to fix glitches in faces & eyes **Show a live preview**: Shows a live preview of the image while it's rendering. Otherwise the image will only be displayed when the image has finished rendering. This is off by default.
* Upscale image by 4x: Uses [RealESRGAN](https://github.com/xinntao/Real-ESRGAN) to upscale the final image to 4x the size (512x512 will be 2048x2048)
**Fix incorrect faces and eyes**: Uses GFPGAN - a face restauration algorithm to fix glitches in faces & eyes.
**Upscale image by 4x**: Uses [RealESRGAN](https://github.com/xinntao/Real-ESRGAN) to upscale the final image to 4x the size (512x512 will be 2048x2048)
* RealESRGAN_x4plus is general-purpose * RealESRGAN_x4plus is general-purpose
* RealESRGAN_x4plus_anime_6B is optimized for anime * RealESRGAN_x4plus_anime_6B is optimized for anime
# Image Modifiers # Image Modifiers
![image](https://user-images.githubusercontent.com/2499585/200583339-783e49a6-8131-4bea-af66-cc311d626156.png)
* Cogwheel icon: Define custom image modifiers ![image](https://user-images.githubusercontent.com/110454200/213927607-9a4323cb-4178-4645-9645-43bf768641f4.png)
* Image style: A pull-down-menu to change the type of icon for the pre-defined modifiers
* Face: Shows faces for the icons in the style of the modifiers **Cogwheel Icon**: To define custom image modifiers.
* Landscapes: Shows landscapes for the icons in the style of the modifiers
* Thumbnail size: A slider to change the size of the clickable modifiers **Image Style**: To change the type of icon for the pre-defined modifiers.
* Image modifiers: These are simply prompt parts, which will be added to the prompt if selected * Face: Shows faces for the icons in the style of the modifiers.
* Landscapes: Shows landscapes for the icons in the style of the modifiers.
**Thumbnail Size**: A slider to change the size of the clickable modifiers.
# Settings # Settings
![image](https://user-images.githubusercontent.com/2499585/202213868-8137c365-5a1b-4dd2-ba9a-a5efb7f06dbd.png) ![image](https://user-images.githubusercontent.com/2499585/202213868-8137c365-5a1b-4dd2-ba9a-a5efb7f06dbd.png)
* Theme: A colour theme for the UI
* Auto-Save Images: Automatically saves images and a .txt-file with the settings to the specified location **Theme**: A color theme for the UI.
* Save Location: A folder for saved images
* Enable Sound: Plays a sound on task completion **Auto-Save Images**: Automatically saves images and a .txt-file with the settings to your specified location.
* Open browser on startup: Starts the default browser on startup with http://localhost:9000/ as URL
* Turbo Mode: generates images faster, but uses an additional 1 GB of GPU memory **Save Location**: A folder for where images will save.
* Use CPU (not GPU): Will force CPU-Rendering (very slow compared to GTX/RTX-Cards)
* Automatically pick the GPUs (experimental): Will select a GPU automatically **Enable Sound**: Plays a sound when a task is completed.
* GPUs to use (experimental): A List of availible GPUs including the CUDA-Id
* Use Full precision: Will use more GPU Memory, required for NVidia GTX 1650/1660. **Open browser on startup**: Starts the default browser on startup with http://localhost:9000/ as URL.
* Auto-Save Settings: Restore Settings on browser load
* Configure: Opens a menu to set which settings are stored **Turbo Mode**: Generates images faster, but uses an additional 1 GB of GPU memory.
* Beta Channel: Get the latest features immediately (but could be less stable). Please restart the program after changing this.
**Use CPU (not GPU)**: Will force CPU-Rendering. Very slow compared to GTX/RTX-Cards.
**Automatically pick the GPUs (experimental)**: Will select a GPU automatically.
**GPUs to use (experimental)**: A List of available GPUs including the CUDA-ID.
**Use Full Precision**: Will use more GPU Memory, required for NVIDIA GTX 1650/1660.
**Auto-Save Settings**: Restores Settings on browser load.
* Configure: Opens a menu to set which settings are stored.
**Beta Channel**: Get the latest features immediately with the downside of decreased stability. Please restart the program after changing this.
# Buttons in the rendered image # Buttons in the rendered image
Note: these are visible if the user hovers with the mouse over a rendered image This is visible if the user hovers with the mouse over a rendered image.
![image](https://user-images.githubusercontent.com/2499585/200590572-b2ef2d74-1a5d-4394-8899-47c8578bedf1.png) ![image](https://user-images.githubusercontent.com/2499585/200590572-b2ef2d74-1a5d-4394-8899-47c8578bedf1.png)
* Seed: The seed this particular image used
* Use as input: Uses the image as Input for img2img **Seed**: The seed this particular image used.
* Download: Downloads the image
* Make Similar images: Enqueues a task using this image for img2img and generates 5 images **Use as Input**: Uses the image as Input for img2img.
* Draw antoher 25 steps: Enquees a task using the the settings of that image and increases the Inference Step Count by 25
* Upscale (only visible if images wasn't already upscaled): Re-Renders the image wtih 4x-Upscaling enabled **Download**: Downloads the image.
* Fix Faces: Fixes incorrect faces and eyes using GFPGAN
**Make Similar Images**: Enqueues a task using this image for img2img and generates 5 images.
**Draw another 25 steps**: Enqueues a task using the the settings of that image and increases the Inference Step Count by 25.
**Upscale**: Re-Renders the image with 4x-Upscaling enabled. Visible if images wasn't already upscaled.
**Fix Faces**: Fixes incorrect faces and eyes using GFPGAN.