Type | |
Stats | 2,532 |
Reviews | (360) |
Published | Mar 20, 2023 |
Base Model | |
Training | Epochs: 20 |
Trigger Words | adolfhitler |
Hash | AutoV2 24752AB128 |
Adolf Diffusion
IMPORTANT NOTES:
>This model is NOT meant to share or glorify any political view.
>Trigger word: "AdolfHitler". Recommended weight: 0.7 and higher
>Trained on b&w pictures, in order to get good results appropriate prompting and model is required.
>The best results achieved with realistic models (Realistic Vision, Deliberate).
>Probably the easiest way is to just use inpaint.
Prompting
>To get images in color use these:
Positive: in color
Negative: monochrome, black and white, b&w, greyscale
>To get rid of blurry background:
Negative: simple background, blurry background
>Trained mostly on upper body pictures, might be hard to get full body.
>If you have problems with getting Adolf's mustache, try setting the LoRa weight higher (or first generate image with low weight and then inpaint face with high weight).
>Eyes might sometimes be a problem, try experimenting with these:
Positive: looking at viewer, detailed eyes etc.
How to make similar model
I know this model is far from being great, but if someone wants to know the workflow, here you go:
1. Find good image and crop it (in my case 44 images). 2. Upscale it x2~x4 (in my case mostly R-ESRGAN). 3. Remove noise using any graphic software. 4. Scale down 5. Train LoRa (using Google Colab).
Also:
>Some of example images were created using ControlNet and img2img
>If you want to use my model anywhere feel free to do so, but it would be nice to put me in credits.
>For any questions you can contact me rafikxd123#7133
Ā