Finderman721 commited on
Commit
59a468d
1 Parent(s): b8e2b62

Big breasts, among us, huge breasts, large knees, sakimichan, juri han, large breasts, beautiful art

Browse files
Files changed (1) hide show
  1. README.md +1 -95
README.md CHANGED
@@ -1,95 +1 @@
1
- ---
2
- language:
3
- - en
4
- tags:
5
- - stable-diffusion
6
- - text-to-image
7
- license: creativeml-openrail-m
8
- inference: false
9
-
10
- ---
11
-
12
- # waifu-diffusion - Diffusion for Weebs
13
-
14
- waifu-diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through fine-tuning.
15
-
16
- # Gradio
17
-
18
- We also support a [Gradio](https://github.com/gradio-app/gradio) web ui with diffusers to run Waifu Diffusion:
19
- [![Open In Spaces](https://camo.githubusercontent.com/00380c35e60d6b04be65d3d94a58332be5cc93779f630bcdfc18ab9a3a7d3388/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f25463025394625413425393725323048756767696e67253230466163652d5370616365732d626c7565)](https://huggingface.co/spaces/hakurei/waifu-diffusion-demo)
20
-
21
- [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1_8wPN7dJO746QXsFnB09Uq2VGgSRFuYE#scrollTo=1HaCauSq546O)
22
-
23
- <img src=https://cdn.discordapp.com/attachments/930559077170421800/1017265913231327283/unknown.png width=40% height=40%>
24
-
25
- [Original PyTorch Model Download Link](https://thisanimedoesnotexist.ai/downloads/wd-v1-2-full-ema.ckpt)
26
-
27
- ## Model Description
28
-
29
- The model originally used for fine-tuning is [Stable Diffusion V1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4), which is a latent image diffusion model trained on [LAION2B-en](https://huggingface.co/datasets/laion/laion2B-en).
30
-
31
- The current model has been fine-tuned with a learning rate of 5.0e-6 for 4 epochs on 56k text-image pairs obtained through Danbooru which all have an aesthetic rating greater than `6.0`.
32
-
33
- **Note:** This project has **no affiliation with Danbooru.**
34
-
35
- ## Training Data & Annotative Prompting
36
-
37
- The data used for fine-tuning has come from a random sample of 56k Danbooru images, which were filtered based on [CLIP Aesthetic Scoring](https://github.com/christophschuhmann/improved-aesthetic-predictor) where only images with an aesthetic score greater than `6.0` were used.
38
-
39
- ## License
40
-
41
- This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
42
- The CreativeML OpenRAIL License specifies:
43
-
44
- 1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
45
- 2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
46
- 3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
47
- [Please read the full license here](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
48
-
49
- ## Downstream Uses
50
-
51
- This model can be used for entertainment purposes and as a generative art assistant.
52
-
53
- ## Example Code
54
-
55
- ```python
56
- import torch
57
- from torch import autocast
58
- from diffusers import StableDiffusionPipeline, DDIMScheduler
59
-
60
- model_id = "hakurei/waifu-diffusion"
61
- device = "cuda"
62
-
63
-
64
- pipe = StableDiffusionPipeline.from_pretrained(
65
- model_id,
66
- torch_dtype=torch.float16,
67
- revision="fp16",
68
- scheduler=DDIMScheduler(
69
- beta_start=0.00085,
70
- beta_end=0.012,
71
- beta_schedule="scaled_linear",
72
- clip_sample=False,
73
- set_alpha_to_one=False,
74
- ),
75
- )
76
- pipe = pipe.to(device)
77
-
78
- prompt = "touhou hakurei_reimu 1girl solo portrait"
79
- with autocast("cuda"):
80
- image = pipe(prompt, guidance_scale=7.5)["sample"][0]
81
-
82
- image.save("reimu_hakurei.png")
83
- ```
84
-
85
- ## Team Members and Acknowledgements
86
-
87
- This project would not have been possible without the incredible work by the [CompVis Researchers](https://ommer-lab.com/).
88
-
89
- - [Anthony Mercurio](https://github.com/harubaru)
90
- - [Salt](https://github.com/sALTaccount/)
91
- - [Sta @ Bit192](https://twitter.com/naclbbr)
92
-
93
- In order to reach us, you can join our [Discord server](https://discord.gg/touhouai).
94
-
95
- [![Discord Server](https://discordapp.com/api/guilds/930499730843250783/widget.png?style=banner2)](https://discord.gg/touhouai)
1
+ Big breasts, among us, huge breasts, large knees, sakimichan, juri han, large breasts, beautiful art