From 3dd26f86a850b704f20ffffff95ac5ac01c7d0af Mon Sep 17 00:00:00 2001 From: chenjian <1435317881@qq.com> Date: Fri, 19 Aug 2022 19:31:57 +0800 Subject: [PATCH] Add release note v2.3.0 --- README.md | 20 +++++++++++++++----- README_ch.md | 24 ++++++++++++++++++------ docs/docs_ch/release.md | 10 ++++++++++ docs/docs_en/release.md | 9 +++++++++ 4 files changed, 52 insertions(+), 11 deletions(-) diff --git a/README.md b/README.md index 0fbb0bb8..fba46692 100644 --- a/README.md +++ b/README.md @@ -29,7 +29,7 @@ English | [简体中文](README_ch.md) ## Introduction and Features - **PaddleHub** aims to provide developers with rich, high-quality, and directly usable pre-trained models. -- **Abundant Pre-trained Models**: 360+ pre-trained models cover the 5 major categories, including Image, Text, Audio, Video, and Industrial application. All of them are free for download and offline usage. +- **Abundant Pre-trained Models**: 360+ pre-trained models cover the 6 major categories, including Wenxin large models, Image, Text, Audio, Video, and Industrial application. All of them are free for download and offline usage. - **No Need for Deep Learning Background**: you can use AI models quickly and enjoy the dividends of the artificial intelligence era. - **Quick Model Prediction**: model prediction can be realized through a few lines of scripts to quickly experience the model effect. - **Model As Service**: one-line command to build deep learning model API service deployment capabilities. @@ -37,22 +37,32 @@ English | [简体中文](README_ch.md) - **Cross-platform**: support Linux, Windows, MacOS and other operating systems. ### Recent updates +- **🔥2022.08.19:** The v2.3.0 version is released, supports Wenxin large models and five text-to-image models based on disco diffusion(dd). + - Support [Wenxin large models API](https://wenxin.baidu.com/moduleApi) for Baidu ERNIE large-scale pre-trained model, including [**ERNIE-ViLG** model](https://aistudio.baidu.com/aistudio/projectdetail/4445016), which supports text-to-image task, and [**ERNIE 3.0 Zeus**](https://aistudio.baidu.com/aistudio/projectdetail/4445054) model, which supports applications such as writing essays, summarization, couplets, question answering, writing novels and completing text. + - Add five text-to-image domain models based on disco diffusion(dd), three for [English](https://aistudio.baidu.com/aistudio/projectdetail/4444984) and two for Chinese. Welcome to enjoy our **ERNIE-ViL**-based Chinese text-to-image module [disco_diffusion_ernievil_base](https://aistudio.baidu.com/aistudio/projectdetail/4444998) in aistudio. - **2022.02.18:** Added Huggingface Org, add spaces and models to the org: [PaddlePaddle Huggingface](https://huggingface.co/PaddlePaddle) -- **2021.12.22**，The v2.2.0 version is released. [1]More than 100 new models released，including dialog, speech, segmentation, OCR, text processing, GANs, and many other categories. The total number of pre-trained models reaches [**【360】**](https://www.paddlepaddle.org.cn/hublist). [2]Add an [indexed file](./modules/README.md) including useful information of pretrained models supported by PaddleHub. [3]Refactor README of pretrained models. -- **2021.05.12:** Add an open-domain dialogue system, i.e., [plato-mini](https://www.paddlepaddle.org.cn/hubdetail?name=plato-mini&en_category=TextGeneration), to make it easy to build a chatbot in wechat with the help of the wechaty, [See Demo](https://github.com/KPatr1ck/paddlehub-wechaty-demo) -- **2021.04.27:** The v2.1.0 version is released. [1] Add supports for five new models, including two high-precision semantic segmentation models based on VOC dataset and three voice classification models. [2] Enforce the transfer learning capabilities for image semantic segmentation, text semantic matching and voice classification on related datasets. [3] Add the export function APIs for two kinds of model formats, i.,e, ONNX and PaddleInference. [4] Add the support for [BentoML](https://github.com/bentoml/BentoML/), which is a cloud native framework for serving deployment. Users can easily serve pre-trained models from PaddleHub by following the [Tutorial notebooks](https://github.com/PaddlePaddle/PaddleHub/blob/release/v2.1/demo/serving/bentoml/cloud-native-model-serving-with-bentoml.ipynb). Also, see this announcement and [Release note](https://github.com/bentoml/BentoML/releases/tag/v0.12.1) from BentoML. (Many thanks to @[parano](https://github.com/parano) @[cqvu](https://github.com/cqvu) @[deehrlic](https://github.com/deehrlic) for contributing this feature in PaddleHub). [5] The total number of pre-trained models reaches **【300】**. -- **2021.02.18:** The v2.0.0 version is released, making model development and debugging easier, and the finetune task is more flexible and easy to use.The ability to transfer learning for visual tasks is fully upgraded, supporting various tasks such as image classification, image coloring, and style transfer; Transformer models such as BERT, ERNIE, and RoBERTa are upgraded to dynamic graphs, supporting Fine-Tune capabilities for text classification and sequence labeling; Optimize the Serving capability, support multi-card prediction, automatic load balancing, and greatly improve performance; the new automatic data enhancement capability Auto Augment can efficiently search for data enhancement strategy combinations suitable for data sets. 61 new word vector models were added, including 51 Chinese models and 10 English models; add 4 image segmentation models, 2 depth models, 7 image generation models, and 3 text generation models, the total number of pre-trained models reaches **【274】**. +- **🔥2021.12.22**，The v2.2.0 version is released. [1]More than 100 new models released，including dialog, speech, segmentation, OCR, text processing, GANs, and many other categories. The total number of pre-trained models reaches [**【360】**](https://www.paddlepaddle.org.cn/hublist). [2]Add an [indexed file](./modules/README.md) including useful information of pretrained models supported by PaddleHub. [3]Refactor README of pretrained models. + - [【more】](./docs/docs_en/release.md) ## Visualization Demo [[More]](./docs/docs_en/visualization.md) [[ModelList]](./modules) + + +### **[Wenxin large models](https://www.paddlepaddle.org.cn/hubdetail?name=ernie_vilg&en_category=TextToImage)** +- Include ERNIE-ViL、ERNIE 3.0 Zeus, supports applications such as text-to-image, writing essays, summarization, couplets, question answering, writing novels and completing text. +
+ +
+ ### **[Computer Vision (212 models)](./modules#Image)**
+ - Many thanks to CopyRight@[PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)、[PaddleDetection](https://github.com/PaddlePaddle/PaddleDetection)、[PaddleGAN](https://github.com/PaddlePaddle/PaddleGAN)、[AnimeGAN](https://github.com/TachibanaYoshino/AnimeGANv2)、[openpose](https://github.com/CMU-Perceptual-Computing-Lab/openpose)、[PaddleSeg](https://github.com/PaddlePaddle/PaddleSeg)、[Zhengxia Zou](https://github.com/jiupinjia/SkyAR)、[PaddleClas](https://github.com/PaddlePaddle/PaddleClas) for the pre-trained models, you can try to train your models with them. diff --git a/README_ch.md b/README_ch.md index 7d6811b6..93018de6 100644 --- a/README_ch.md +++ b/README_ch.md @@ -30,7 +30,7 @@ ## 简介与特性 - 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 -- **【模型种类丰富】**: 涵盖CV、NLP、Audio、Video、工业应用主流五大品类的 **360+** 预训练模型，全部开源下载，离线可运行 +- **【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 - **【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 - **【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 - **【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 @@ -38,18 +38,30 @@ - **【跨平台兼容性】**：可运行于Linux、Windows、MacOS等多种操作系统 ## 近期更新 -- **2021.12.22**，发布v2.2.0版本。【1】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+】**](https://www.paddlepaddle.org.cn/hublist)；【2】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；【3】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。 -- **2021.05.12**，新增轻量级中文对话模型[plato-mini](https://www.paddlepaddle.org.cn/hubdetail?name=plato-mini&en_category=TextGeneration)，可以配合使用wechaty实现微信闲聊机器人，[参考demo](https://github.com/KPatr1ck/paddlehub-wechaty-demo) -- **2021.04.27**，发布v2.1.0版本。【1】新增基于VOC数据集的高精度语义分割模型2个，语音分类模型3个。【2】新增图像语义分割、文本语义匹配、语音分类等相关任务的Fine-Tune能力以及相关任务数据集;完善部署能力：【3】新增ONNX和PaddleInference等模型格式的导出功能。【4】新增[BentoML](https://github.com/bentoml/BentoML) 云原生服务化部署能力，可以支持统一的多框架模型管理和模型部署的工作流，[详细教程](https://github.com/PaddlePaddle/PaddleHub/blob/release/v2.1/demo/serving/bentoml/cloud-native-model-serving-with-bentoml.ipynb). 更多内容可以参考BentoML 最新 v0.12.1 [Releasenote](https://github.com/bentoml/BentoML/releases/tag/v0.12.1).（感谢@[parano](https://github.com/parano) @[cqvu](https://github.com/cqvu) @[deehrlic](https://github.com/deehrlic)）的贡献与支持。【5】预训练模型总量达到[**【300】**](https://www.paddlepaddle.org.cn/hublist)个。 -- **2021.02.18**，发布v2.0.0版本，【1】模型开发调试更简单，finetune接口更加灵活易用。视觉类任务迁移学习能力全面升级，支持[图像分类](./demo/image_classification/README.md)、[图像着色](./demo/colorization/README.md)、[风格迁移](./demo/style_transfer/README.md)等多种任务；BERT、ERNIE、RoBERTa等Transformer类模型升级至动态图，支持[文本分类](./demo/text_classification/README.md)、[序列标注](./demo/sequence_labeling/README.md)的Fine-Tune能力；【2】优化服务化部署Serving能力，支持多卡预测、自动负载均衡，性能大幅度提升；【3】新增自动数据增强能力[Auto Augment](./demo/autoaug/README.md)，能高效地搜索适合数据集的数据增强策略组合。【4】新增[词向量模型](./modules/text/embedding)61个，其中包含中文模型51个，英文模型10个；新增[图像分割](./modules/thirdparty/image/semantic_segmentation)模型4个、[深度模型](./modules/thirdparty/image/depth_estimation)2个、[图像生成](./modules/thirdparty/image/Image_gan/style_transfer)模型7个、[文本生成](./modules/thirdparty/text/text_generation)模型3个。【5】预训练模型总量达到[**【274】**](https://www.paddlepaddle.org.cn/hublist) 个。 -- [More](./docs/docs_ch/release.md) +- **🔥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。 + - 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))。 + - 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))。 +- **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)。 + +- **🔥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+】； + - 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； + - 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。 +- [More](./docs/docs_ch/release.md) ## **精品模型效果展示[【更多】](./docs/docs_ch/visualization.md)[【模型库】](./modules/README_ch.md)** +### **[文心大模型](https://www.paddlepaddle.org.cn/hubdetail?name=ernie_vilg&en_category=TextToImage)** +- 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。 +
+ +
+ + ### **[图像类（212个）](./modules/README_ch.md#图像)** - 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
diff --git a/docs/docs_ch/release.md b/docs/docs_ch/release.md index b7e82c9d..8eebd720 100755 --- a/docs/docs_ch/release.md +++ b/docs/docs_ch/release.md @@ -1,5 +1,15 @@ # 更新历史 +## `v2.3.0` + +### 【1、支持文图生成新场景】 + - 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3个，中文模型2个，其中中文文图生成模型[disco_diffusion_ernievil_base](https://aistudio.baidu.com/aistudio/projectdetail/4444998)基于百度自研多模态模型**ERNIE-ViL**开发，欢迎体验。 + +### 【2、支持文心大模型API调用】 + - 新增对文心大模型[**ERNIE-ViLG**](https://aistudio.baidu.com/aistudio/projectdetail/4445016)的API调用，支持文图生成任务。 + - 新增对文心大模型[**ERNIE 3.0 Zeus**](https://aistudio.baidu.com/aistudio/projectdetail/4445054)的API调用，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。 + + ## `v2.1.0` ### 【1、版本迭代】 diff --git a/docs/docs_en/release.md b/docs/docs_en/release.md index 407d95ca..781ecdf6 100755 --- a/docs/docs_en/release.md +++ b/docs/docs_en/release.md @@ -1,5 +1,14 @@ # Release Note +## `v2.3.0` + +### [1、Support text-to-image domain model] + - Add five text-to-image domain models based on disco diffusion, in which three models are for English and two for Chinese. Especially, Chinese text-to-image model [disco_diffusion_ernievil_base](https://aistudio.baidu.com/aistudio/projectdetail/4444998) is based on Baidu **ERNIE-ViL**，welcome to experience. + +### 【2、Support Wenxin large models API】 + - Add api call for [**ERNIE-ViLG**](https://aistudio.baidu.com/aistudio/projectdetail/4445016) model, which supports text-to-image task。 + - Add api call for [**ERNIE 3.0 Zeus**](https://aistudio.baidu.com/aistudio/projectdetail/4445054) model, which supports applications such as writing essays, summarization, couplets, question answering, writing novels and completing text. + ## `v2.1.0` ### [ 1. Improvements] -- GitLab