Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-sa-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-sa-4.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
pipeline_tag: text-generation
|
| 6 |
+
tags:
|
| 7 |
+
- code
|
| 8 |
---
|
| 9 |
+
# MagicPrompt_SD_V1
|
| 10 |
+
|
| 11 |
+
This is a Prompt Generator likes [Gustavosta/MagicPrompt-Stable-Diffusion](https://huggingface.co/Gustavosta/MagicPrompt-Stable-Diffusion)!
|
| 12 |
+
|
| 13 |
+
But I wash the origin prompts data, and trains a powerful model to generate prompt for [魔导绪论](https://magic-tag.netlify.app/)
|
| 14 |
+
|
| 15 |
+
It's using Paddle to handle the training and other things. Not PyTorch or Tensorsflow.
|
| 16 |
+
|
| 17 |
+
There's the result I get form this model:
|
| 18 |
+
|
| 19 |
+
- You can use CPU to run the model! But GPU 10x faster then CPU 🚀.
|
| 20 |
+
- CPU (about 1200ms/per ) | GPU (about 200ms/per 🚀 )
|
| 21 |
+
- You can add some change easier passing some params.
|
| 22 |
+
|
| 23 |
+
## 📕 Using example is here
|
| 24 |
+
|
| 25 |
+
[飞桨仓库](https://aistudio.baidu.com/aistudio/projectdetail/5116158?contributionType=1)
|
| 26 |
+
|
| 27 |
+
You can wrapper a FastAPI or Flask to easily deploy it to your server
|