Skip to content

Commit 5699dfb

Browse files
authored
big modle readme (PaddlePaddle#2117)
1 parent 751f30b commit 5699dfb

File tree

4 files changed

+86
-16
lines changed

4 files changed

+86
-16
lines changed

README.md

+29-3
Original file line numberDiff line numberDiff line change
@@ -39,12 +39,38 @@ English | [简体中文](README_ch.md)
3939
<a name="demos"></a>
4040
## 🌈Visualization Demo
4141

42-
#### 🏜️ [Text-to-Image Models](https://www.paddlepaddle.org.cn/hubdetail?name=ernie_vilg&en_category=TextToImage)
43-
- Include ERNIE-ViLG, ERNIE-ViL, ERNIE 3.0 Zeus, supports applications such as text-to-image, writing essays, summarization, couplets, question answering, writing novels and completing text.
42+
43+
44+
45+
46+
#### 🏜️ [Text-to-Image Models](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image)
4447
<div align="center">
45-
<img src="https://user-images.githubusercontent.com/22424850/185588578-e2d1216b-e797-458d-bc6b-0ccb8e1bd1b9.png" width = "80%" />
48+
<table>
49+
<tr>
50+
<td><img src="https://user-images.githubusercontent.com/59186797/200235049-fefa7642-6c4c-4f93-bd84-3b36a8a80595.gif" width = "100%"></td>
51+
<td><img src="https://user-images.githubusercontent.com/59186797/200244625-77310db8-c9b2-4293-8fe9-c9aae27ee462.gif" width = "80%"></td>
52+
<td><img src="https://user-images.githubusercontent.com/59186797/200245387-daaf576d-8224-4937-82b8-27e31ee2df16.gif" width = "100%"></td>
53+
<tr>
54+
<tr>
55+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/ernie_vilg">Wenxin Big Moels</a></td>
56+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion">Stable_Diffusion series</a></td>
57+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/disco_diffusion_ernievil_base">Disco Diffusion series</a></td>
58+
59+
<tr>
60+
61+
<tr>
62+
<td align="center">Include ERNIE-ViLG, ERNIE-ViL, ERNIE 3.0 Zeus, supports applications such as text-to-image, writing essays, summarization, couplets, question answering, writing novels and completing text。</td>
63+
<td align="center">Supports functions such as text_to_image, image_to_image, inpainting, ACGN external service, etc.</td>
64+
<td align="center">Support Chinese and English input</td>
65+
66+
<tr>
67+
68+
</table>
4669
</div>
4770

71+
72+
73+
4874
#### 👓 [Computer Vision Models](./modules#Image)
4975
<div align="center">
5076
<img src="./docs/imgs/Readme_Related/Image_all.gif" width = "530" height = "400" />

README_ch.md

+35-12
Original file line numberDiff line numberDiff line change
@@ -30,22 +30,23 @@
3030

3131
## 简介与特性
3232
- PaddleHub旨在为开发者提供丰富的、高质量的、直接可用的预训练模型
33-
- **【模型种类丰富】**: 涵盖大模型、CV、NLP、Audio、Video、工业应用主流六大品类的 **360+** 预训练模型,全部开源下载,离线可运行
33+
- **【模型种类丰富】**: 涵盖大模型、CV、NLP、Audio、Video、工业应用主流六大品类的 **400+** 预训练模型,全部开源下载,离线可运行
3434
- **【超低使用门槛】**:无需深度学习背景、无需数据与训练过程,可快速使用AI模型
3535
- **【一键模型快速预测】**:通过一行命令行或者极简的Python API实现模型调用,可快速体验模型效果
3636
- **【一键模型转服务化】**:一行命令,搭建深度学习模型API服务化部署能力
3737
- **【跨平台兼容性】**:可运行于Linux、Windows、MacOS等多种操作系统
3838

3939
## 近期更新
40-
- **🔥2022.08.19:** 发布v2.3.0版本新增[文心大模型](https://wenxin.baidu.com/)和disco diffusion(dd)系列文图生成模型。
41-
- 支持对[文心大模型API](https://wenxin.baidu.com/moduleApi)的调用, 包括 文图生成模型**ERNIE-ViLG**([体验Demo](https://aistudio.baidu.com/aistudio/projectdetail/4445016)), 以及支持写作文、写文案、写摘要、对对联、自由问答、写小说、补全文本等多个应用的语言模型**ERNIE 3.0 Zeus**([体验Demo](https://aistudio.baidu.com/aistudio/projectdetail/4445054))。
42-
- 新增基于disco diffusion技术的文图生成dd系列模型5个,其中英文模型([体验Demo](https://aistudio.baidu.com/aistudio/projectdetail/4444984))3个,中文模型2个。欢迎点击链接在aistudio上进行体验基于**ERNIE-ViL**开发的中文文图生成模型disco_diffusion_ernievil_base([体验Demo](https://aistudio.baidu.com/aistudio/projectdetail/4444998))。
43-
- **2022.02.18:** 加入Huggingface,创建了PaddlePaddle的空间并上传了模型: [PaddlePaddle Huggingface](https://huggingface.co/PaddlePaddle)
40+
- **🔥2022.10.20:** 发布v2.3.1版本新增Stable_Diffusion系列模型和超分模型
41+
- 支持[文生图](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion)[图生图](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion_img2img)[图修复](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion_inpainting)[二次元专属waifu](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion_waifu)等4个模型。
42+
- 基于 [SwinIR-L](https://www.paddlepaddle.org.cn/hubdetail?name=swinir_l_real_sr_x4&en_category=ImageEditing) 的 4 倍现实图像超分辨率模型
43+
- **🔥2022.08.19:** 发布v2.3.0版本新增[文心大模型](https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/ernie_vilg)[disco diffusion(dd)](https://www.paddlepaddle.org.cn/hubdetail?name=disco_diffusion_ernievil_base&en_category=TextToImage)系列文图生成模型。
44+
- 支持对[文心大模型API](https://wenxin.baidu.com/moduleApi)的调用, 包括 文图生成模型ERNIE-ViLG, 以及支持写作文、写文案、写摘要、对对联、自由问答、写小说、补全文本等多个应用的语言模型ERNIE 3.0 Zeus
45+
- 新增基于disco diffusion技术的文图生成dd系列模型([免费GPU体验Demo](https://aistudio.baidu.com/aistudio/projectdetail/4462918))。
46+
- **2022.02.18:** 加入Huggingface,创建了PaddlePaddle的可视化空间并上传了模型: [PaddlePaddle Huggingface](https://huggingface.co/PaddlePaddle)
4447

45-
- **🔥2021.12.22**,发布v2.2.0版本新增[预训练模型库官网](https://www.paddlepaddle.org.cn/hublist)
46-
- 新增100+高质量模型,涵盖对话、语音处理、语义分割、文字识别、文本处理、图像生成等多个领域,预训练模型总量达到【360+】;
47-
- 新增模型[检索列表](./modules/README_ch.md),包含模型名称、网络、数据集和使用场景等信息,快速定位用户所需的模型;
48-
- 模型文档排版优化,呈现数据集、指标、模型大小等更多实用信息。
48+
- **🔥2021.12.22**,发布v2.2.0版本新增[预训练模型库官网](https://www.paddlepaddle.org.cn/hublist),新增100+高质量模型,涵盖对话、语音处理、语义分割、文字识别、文本处理、图像生成等多个领域,预训练模型总量达到【360+】;
49+
4950

5051

5152
- [More](./docs/docs_ch/release.md)
@@ -54,13 +55,35 @@
5455

5556
## **精品模型效果展示[【更多】](./docs/docs_ch/visualization.md)[【模型库】](./modules/README_ch.md)**
5657

57-
### **[文心大模型](https://www.paddlepaddle.org.cn/hubdetail?name=ernie_vilg&en_category=TextToImage)**
58-
- 包含大模型ERNIE-ViL、ERNIE 3.0 Zeus, 支持文图生成、写作文、写文案、写摘要、对对联、自由问答、写小说、补全文本等多个应用。
58+
### **[大模型(10个)](./modules/README_ch.md#图像)**
5959
<div align="center">
60-
<img src="https://user-images.githubusercontent.com/22424850/185588578-e2d1216b-e797-458d-bc6b-0ccb8e1bd1b9.png" width = "80%" />
60+
<table>
61+
<tr>
62+
<td><img src="https://user-images.githubusercontent.com/59186797/200235049-fefa7642-6c4c-4f93-bd84-3b36a8a80595.gif" width = "100%"></td>
63+
<td><img src="https://user-images.githubusercontent.com/59186797/200244625-77310db8-c9b2-4293-8fe9-c9aae27ee462.gif" width = "90%"></td>
64+
<td><img src="https://user-images.githubusercontent.com/59186797/200245387-daaf576d-8224-4937-82b8-27e31ee2df16.gif" width = "100%"></td>
65+
<tr>
66+
<tr>
67+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/ernie_vilg">文心大模型</a></td>
68+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/stable_diffusion">Stable_Diffusion系列模型</a></td>
69+
<td align="center"><a href="https://github.com/PaddlePaddle/PaddleHub/tree/develop/modules/image/text_to_image/disco_diffusion_ernievil_base">Disco Diffusion系列模型</a></td>
70+
71+
<tr>
72+
73+
<tr>
74+
<td align="center">支持文图生成、写作文、写文案、写摘要、对对联、自由问答、写小说、补全文本等多个应用。</td>
75+
<td align="center">支持文生图、图生图、图修复、二次元专属waifu等功能</td>
76+
<td align="center">支持中英输入</td>
77+
78+
<tr>
79+
80+
</table>
6181
</div>
6282

6383

84+
85+
86+
6487
### **[图像类(212个)](./modules/README_ch.md#图像)**
6588
- 包括图像分类、人脸检测、口罩检测、车辆检测、人脸/人体/手部关键点检测、人像分割、80+语言文本识别、图像超分/上色/动漫化等
6689
<div align="center">

modules/README.md

+8
Original file line numberDiff line numberDiff line change
@@ -286,6 +286,14 @@ English | [简体中文](README_ch.md)
286286
|[MiDaS_Large](image/depth_estimation/MiDaS_Large)|-|3D Movies, WSVD, ReDWeb, MegaDepth|| [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/PaddlePaddle/MiDaS_Large) |
287287
|[MiDaS_Small](image/depth_estimation/MiDaS_Small)|-|3D Movies, WSVD, ReDWeb, MegaDepth, etc.|| [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/PaddlePaddle/MiDaS_Small) |
288288

289+
- ### Text_to_Image
290+
291+
|module|Network|Dataset|Introduction|Huggingface Spaces Demo|
292+
|--|--|--|--|--|
293+
|[disco_diffusion_clip_rn101](image/text_to_image/disco_diffusion_clip_rn101)|-|Open domain multi round dataset|text_to_image|
294+
|[ernie_vilg](image/text_to_image/ernie_vilg)|-|Open domain multi round dataset|text_to_image|[![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/PaddlePaddle/ERNIE-ViLG) |
295+
|[stable_diffusion_img2img](image/text_to_image/stable_diffusion_img2img)|-|Open domain multi round dataset|img2img|
296+
289297
## Text
290298
- ### Text Generation
291299

modules/README_ch.md

+14-1
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
简体中文 | [English](README.md)
22

33
# 目录
4-
|[图像](#图像)212个|[文本](#文本) (130个)|[语音](#语音) (15个)|[视频](#视频) (8个)|[工业应用](#工业应用) (1个)|
4+
|[图像](#图像)222个|[文本](#文本) (130个)|[语音](#语音) (15个)|[视频](#视频) (8个)|[工业应用](#工业应用) (1个)|
55
|--|--|--|--|--|
66
|[图像分类](#图像分类) (108)|[文本生成](#文本生成) (17)| [声音克隆](#声音克隆) (2)|[视频分类](#视频分类) (5)| [表针识别](#表针识别) (1)|
77
|[图像生成](#图像生成) (26)|[词向量](#词向量) (62)|[语音合成](#语音合成) (5)|[视频修复](#视频修复) (1)|-|
@@ -13,6 +13,7 @@
1313
|[实例分割](#实例分割) (1)|[词法分析](#词法分析) (2)|-|-|-|
1414
|[目标检测](#目标检测) (13)|[标点恢复](#标点恢复) (1)|-|-|-|
1515
|[深度估计](#深度估计) (2)|[文本审核](#文本审核) (3)|-|-|-|
16+
|[文生图](#文生图) (10)|-|-|-|-|
1617

1718
## 图像
1819
- ### 图像分类
@@ -285,6 +286,18 @@
285286
|[MiDaS_Large](image/depth_estimation/MiDaS_Large)|-|3D Movies, WSVD, ReDWeb, MegaDepth||
286287
|[MiDaS_Small](image/depth_estimation/MiDaS_Small)|-|3D Movies, WSVD, ReDWeb, MegaDepth, etc.||
287288

289+
- ### 文生图
290+
291+
|module|网络|数据集|简介|
292+
|--|--|--|--|
293+
|[disco_diffusion_clip_rn101](image/text_to_image/disco_diffusion_clip_rn101)|-|开放域多轮数据集|文生图|
294+
|[ernie_vilg](image/text_to_image/ernie_vilg)|-|开放域多轮数据集|文生图|
295+
|[stable_diffusion_img2img](image/text_to_image/stable_diffusion_img2img)|-|开放域多轮数据集|图生图|
296+
297+
298+
299+
300+
288301
## 文本
289302
- ### 文本生成
290303

0 commit comments

Comments
 (0)