Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,68 @@ sdk: static
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
+
## Who We Are
|
11 |
+
|
12 |
+
We are a team focused on automating the training process of stable diffusion models (LoRA, Lycoris, Pivotal Tuning, etc.).
|
13 |
+
|
14 |
+
Our goal is to automate the entire pipeline, including data acquisition, data filtering, training, step selection, and platform deployment, **saving manpower and optimizing model training quality to the maximum**.
|
15 |
+
|
16 |
+
Our team comprises a Ph.D. in Software Engineering, a Ph.D. candidate in Computer Vision, professionals in art and design, and several AI waifu enthusiasts.
|
17 |
+
|
18 |
+
We are a purely non-profit team, and all our work is completely open, without any form of charge.
|
19 |
+
|
20 |
+
## Our Achievements
|
21 |
+
|
22 |
+
We have conducted multiple iterations on Pivotal Tuning training technology (training with a LoRA and one or more pt files). The progress is documented in the following technical blogs:
|
23 |
+
* [Overview of v1.4 Training Automation](https://civitai.com/articles/2064/2023-8-31-release-of-v14-training-automation-process)
|
24 |
+
* [Result Analysis of v1.4 Training Automation](https://civitai.com/articles/2479/2023-10-7-survey-of-v14-training-automation-and-planning-of-version-v15)
|
25 |
+
* v1.5 Automation is coming... :)
|
26 |
+
|
27 |
+
Models we have trained or hosted can be found at:
|
28 |
+
* [CyberHarem - Huggingface](https://huggingface.co/CyberHarem), including datasets used for training
|
29 |
+
* [narugo1992 - Civitai](https://civitai.com/user/narugo1992), where only a few models are selected for publication to avoid offense to those who insist on manually training models, lol
|
30 |
+
|
31 |
+
Quantitative analysis from the blog "Result Analysis of v1.4 Training Automation" shows that **v1.4 training automation has achieved a quite impressive level in both quality and quantity**, but there's still room for further improvement. This is an ongoing effort for v1.5 and future versions.
|
32 |
+
|
33 |
+
## Our Technical Outputs
|
34 |
+
|
35 |
+
### dghs-imgutils
|
36 |
+
|
37 |
+
Project Link: https://github.com/deepghs/imgutils
|
38 |
+
|
39 |
+
Project Documentation: https://deepghs.github.io/imgutils/main/index.html
|
40 |
+
|
41 |
+
**This is a library for various common operations on anime images**, including but not limited to:
|
42 |
+
|
43 |
+
* Tachie (Difference) Detection and Clustering
|
44 |
+
* Contrastive Character Image Pretraining
|
45 |
+
* Object Detection
|
46 |
+
* Edge Detection / Lineart Generation
|
47 |
+
* Monochrome Image Detection
|
48 |
+
* Image Rating
|
49 |
+
* Truncated Image Check
|
50 |
+
* Image Tagging
|
51 |
+
* Character Extraction
|
52 |
+
|
53 |
+
Check out the documentation for more features.
|
54 |
+
|
55 |
+
### Waifuc
|
56 |
+
|
57 |
+
Project Link: https://github.com/deepghs/waifuc
|
58 |
+
|
59 |
+
A **data pipeline framework based on dghs-imgutils**, supporting:
|
60 |
+
* Fast data retrieval (local disk, danbooru, pixiv, zerochan, etc.)
|
61 |
+
* Swift data filtering (comic exclusion, monochrome image exclusion, multi-character image exclusion, irrelevant character exclusion, etc.)
|
62 |
+
* Rapid data saving (local, cloud; with metadata, saved in stable diffusion dataset format, etc.)
|
63 |
+
* Quick building of processing pipelines (connecting multiple aforementioned stages)
|
64 |
+
|
65 |
+
*Note: This tool is currently a work in progress, although it's in use. It hasn't been released on PyPI and lacks comprehensive documentation. These aspects will be addressed soon.*
|
66 |
+
|
67 |
+
### Model Zoo
|
68 |
+
|
69 |
+
We manage our models and datasets on Huggingface: https://huggingface.co/deepghs
|
70 |
+
|
71 |
+
### Anything More?
|
72 |
+
|
73 |
+
In fact, our plans go beyond what's mentioned here. Other tools are continuously improving and will soon be released. Stay tuned!
|
74 |
+
|