Update README.md
Browse files- README.md +24 -22
- money-wallet.gif +0 -0
- preview1.png +0 -0
- preview2.png +0 -0
- preview3.png +0 -0
README.md
CHANGED
|
@@ -1,15 +1,6 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: agpl-3.0
|
| 3 |
-
language:
|
| 4 |
-
- en
|
| 5 |
-
- vi
|
| 6 |
-
pipeline_tag: text-to-image
|
| 7 |
-
tags:
|
| 8 |
-
- art
|
| 9 |
-
---
|
| 10 |
# Waifumake (●'◡'●) AI Art model.
|
| 11 |
|
| 12 |
-

|
| 13 |
|
| 14 |
A single student training an AI model that generates art.
|
| 15 |
|
|
@@ -23,41 +14,50 @@ A single student training an AI model that generates art.
|
|
| 23 |
- You should use it as a supportive tool for creating works of art, and not rely on it completely.
|
| 24 |
- It can be used as a tool for upscaling or rendering anime-style images from 3D modeling software (Blender).
|
| 25 |
- Create an image from a sketch you created from a pure drawing program. (MS Paint)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
## Training:
|
| 28 |
- **Data**: The model is trained based on a database of various sources from the Internet provided by my friend and images created by another AI.
|
| 29 |
-
- **Schedule**: Euler Ancestral.
|
| 30 |
- **Optimizer**: AdamW.
|
| 31 |
-
- **Precision**: BF16
|
| 32 |
- **Hardware**: Google Colaboratory Pro - NVIDIA A100 40GB VRAM.
|
| 33 |
|
| 34 |
-
## Limitations
|
| 35 |
-
### **Limitations:**
|
| 36 |
- Loss of detail, errors, bad human-like (six-fingered hand) details, deformation, blurring, and unclear images are inevitable.
|
| 37 |
- Complex tasks cannot be handled.
|
| 38 |
-
-
|
| 39 |
-
- Content may not be appropriate for all ages: As it is trained on data that includes adult content, the generated images may contain content not suitable for children (depending on your country there will be a specific regulation about it). If you do not want to appear adult content, make sure you have additional safety measures in place, such as adding "NSFW" to the negative prompt.
|
| 40 |
- The results generated by the model are considered impressive. But unfortunately, currently, it only supports the English language, to use multilingual, consider using third-party translation programs.
|
| 41 |
- The model is trained on the `Danbooru` and `Nai` tagging system, so the long text may result in poor results.
|
| 42 |
-
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
Want to support me? Thank you, please help me make it better. ❤️
|
| 47 |
|
| 48 |
## <p style="color:red">⚠️Prohibited behaviors:<p>
|
| 49 |
- Using for political, terrorist, subversive, racist, disrespectful of law, and lawless purposes.
|
| 50 |
- Stealing, copying, or reproducing someone else's work without permission for commercial or malicious purposes.
|
| 51 |
- Spreading false information.
|
| 52 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
## Special Thank:
|
| 54 |
This wouldn't have happened if they hadn't made a breakthrough.
|
| 55 |
- [Runwayml](https://huggingface.co/runwayml/): Base model.
|
| 56 |
- [d8ahazard](https://github.com/d8ahazard/.sd_dreambooth_extension) : Dreambooth.
|
| 57 |
- [Automatic1111](https://github.com/AUTOMATIC1111/) : Web UI.
|
| 58 |
- [Mikubill](https://github.com/Mikubill/): Where my ideas started.
|
| 59 |
-
-
|
| 60 |
- Novel AI: Dataset images. An AI made me thousands of pictures without worrying about copyright or dispute.
|
|
|
|
| 61 |
- My friend and others.
|
| 62 |
- YOU! Yes, is you 🫵
|
| 63 |
|
|
@@ -69,4 +69,6 @@ If any part of the model does not comply with the terms of the GNU General Publi
|
|
| 69 |
|
| 70 |
We will not be held responsible for any legal issues you cause.
|
| 71 |
|
|
|
|
|
|
|
| 72 |
# Have fun with your waifu! (●'◡'●)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Waifumake (●'◡'●) AI Art model.
|
| 2 |
|
| 3 |
+

|
| 4 |
|
| 5 |
A single student training an AI model that generates art.
|
| 6 |
|
|
|
|
| 14 |
- You should use it as a supportive tool for creating works of art, and not rely on it completely.
|
| 15 |
- It can be used as a tool for upscaling or rendering anime-style images from 3D modeling software (Blender).
|
| 16 |
- Create an image from a sketch you created from a pure drawing program. (MS Paint)
|
| 17 |
+
- The `masterpiece` and `best quality` tags are not necessary, as it sometimes leads to contradictory results, but if it is distorted or discolored, add them now.
|
| 18 |
+
|
| 19 |
+
## Preview:
|
| 20 |
+

|
| 21 |
+

|
| 22 |
+

|
| 23 |
|
| 24 |
## Training:
|
| 25 |
- **Data**: The model is trained based on a database of various sources from the Internet provided by my friend and images created by another AI.
|
| 26 |
+
- **Schedule**: Euler Ancestral Discrete.
|
| 27 |
- **Optimizer**: AdamW.
|
| 28 |
+
- **Precision**: BF16.
|
| 29 |
- **Hardware**: Google Colaboratory Pro - NVIDIA A100 40GB VRAM.
|
| 30 |
|
| 31 |
+
## **Limitations:**
|
|
|
|
| 32 |
- Loss of detail, errors, bad human-like (six-fingered hand) details, deformation, blurring, and unclear images are inevitable.
|
| 33 |
- Complex tasks cannot be handled.
|
| 34 |
+
- ⚠️Content may not be appropriate for all ages: As it is trained on data that includes adult content, the generated images may contain content not suitable for children (depending on your country there will be a specific regulation about it). If you do not want to appear adult content, make sure you have additional safety measures in place, such as adding "nsfw" to the negative prompt.
|
|
|
|
| 35 |
- The results generated by the model are considered impressive. But unfortunately, currently, it only supports the English language, to use multilingual, consider using third-party translation programs.
|
| 36 |
- The model is trained on the `Danbooru` and `Nai` tagging system, so the long text may result in poor results.
|
| 37 |
+
- Dark, grayscale, white balance loss. The fix is to use image editing software like Photoshop. In the future, I need a more colorful dataset.
|
| 38 |
+
- My amount of money: 0 USD =((.
|
| 39 |
+
|
| 40 |
+

|
|
|
|
| 41 |
|
| 42 |
## <p style="color:red">⚠️Prohibited behaviors:<p>
|
| 43 |
- Using for political, terrorist, subversive, racist, disrespectful of law, and lawless purposes.
|
| 44 |
- Stealing, copying, or reproducing someone else's work without permission for commercial or malicious purposes.
|
| 45 |
- Spreading false information.
|
| 46 |
|
| 47 |
+
## **Desires:**
|
| 48 |
+
As it is a version made only by myself and my small associates, the model will not be perfect and may differ from what people expect. Any contributions from everyone will be respected.
|
| 49 |
+
|
| 50 |
+
Want to support me? Thank you, please help me make it better. ❤️
|
| 51 |
+
|
| 52 |
## Special Thank:
|
| 53 |
This wouldn't have happened if they hadn't made a breakthrough.
|
| 54 |
- [Runwayml](https://huggingface.co/runwayml/): Base model.
|
| 55 |
- [d8ahazard](https://github.com/d8ahazard/.sd_dreambooth_extension) : Dreambooth.
|
| 56 |
- [Automatic1111](https://github.com/AUTOMATIC1111/) : Web UI.
|
| 57 |
- [Mikubill](https://github.com/Mikubill/): Where my ideas started.
|
| 58 |
+
- Chat-GPT: Help me do crazy things that I thought I would never do.
|
| 59 |
- Novel AI: Dataset images. An AI made me thousands of pictures without worrying about copyright or dispute.
|
| 60 |
+
- Danbooru: Help me write the correct tag.
|
| 61 |
- My friend and others.
|
| 62 |
- YOU! Yes, is you 🫵
|
| 63 |
|
|
|
|
| 69 |
|
| 70 |
We will not be held responsible for any legal issues you cause.
|
| 71 |
|
| 72 |
+
Don't forget me.
|
| 73 |
+
|
| 74 |
# Have fun with your waifu! (●'◡'●)
|
money-wallet.gif
ADDED
|
preview1.png
ADDED
|
preview2.png
ADDED
|
preview3.png
ADDED
|