File size: 1,182 Bytes
8025393
 
 
08cafa6
301c462
 
 
 
08cafa6
 
fdd22ef
f0821e6
 
223fba4
b8768c6
 
 
 
f5a2262
b8768c6
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
---
license: creativeml-openrail-m
---
Put 'mkmk woman' in prompt to generate Marco's pictures
<img src="https://huggingface.co./AkiKagura/mkgen-diffusion/blob/main/samples/0.png" width="256" height="256"/>
<img src="https://huggingface.co./AkiKagura/mkgen-diffusion/blob/main/samples/1.png" width="256" height="256"/>
<img src="https://huggingface.co./AkiKagura/mkgen-diffusion/blob/main/samples/2.png" width="256" height="256"/>
<img src="https://huggingface.co./AkiKagura/mkgen-diffusion/blob/main/samples/3.png" width="256" height="256"/>

A stable diffusion model used to generate Marco's pictures

Based on runwayml/stable-diffusion-v1-5 trained by Dreambooth

This package contains diffusers model and ckpt model.(Attention: these two models are trained in different ways so they may not be performing the same)
Trained on 39 pics, 3000 steps

To use diffusers model, set code like this:
`pipe = StableDiffusionPipeline.from_pretrained("AkiKagura/mkgen-diffusion")`
You could also try it on https://colab.research.google.com/drive/1xvB8DH-KL_GfgfpH0XyVlQNy-6rywAcw?usp=share_link

To use ckpt model: download the ckpt file in this package and use it in stable diffusion projects