From 5690f6181858e5e3cb8d6a499fec657155218a74 Mon Sep 17 00:00:00 2001 From: Joker1718 <82336164+Joker1718@users.noreply.github.com> Date: Wed, 3 May 2023 09:05:25 +0800 Subject: [PATCH 1/8] [Translation] Update on README_en.md Doing a community job for English developers and updating the translations for README_en.md. If you have any questions or problems, feel free to ask. :+1: --- README_en.md | 52 +++++++++++++++++++--------------------------------- 1 file changed, 19 insertions(+), 33 deletions(-) diff --git a/README_en.md b/README_en.md index efebb667c84b..914c33fb621f 100644 --- a/README_en.md +++ b/README_en.md @@ -1,46 +1,32 @@ -[简体中文🀄](./README.md) | **English**🌎 -

- -

+[简体中文🀄](./README.MD) | **English🌎** ------------------------------------------------------------------------------------------- +

-

- - - - - - - - - -

+* * * * * -

- Features | - Installation | - Quick Start | - API Reference | - Community -

+

-**PaddleNLP** is an *easy-to-use* and *powerful* NLP library with **Awesome** pre-trained model zoo, supporting wide-range of NLP tasks from research to industrial applications. +

Features | Installation | Quick Start | API Documentation | Community -## News 📢 +**PaddleNLP** is a natural language processing development library that is both **easy to use** and **powerful**. It aggregates high-quality pre-trained models in the industry and provides a **plug-and-play** development experience, covering a model library for various NLP scenarios. With practical examples from industry practices, PaddleNLP can meet the needs of developers who require **flexible customization**. -* 🔥 **Latest Features** - * 📃 Release **[UIE-X](./applications/information_extraction)**, an universal information extraction model that supports both document and text inputs. - * ❣️Release **[Opinion Mining and Sentiment Analysis Models](./applications/sentiment_analysis/unified_sentiment_extraction)** based on UIE, including abilities of sentence-level and aspect-based sentiment classification, attribute extraction, opinion extraction, attribute aggregation and implicit opinion extraction. -* **2022.9.6 [PaddleNLPv2.4](https://github.com/PaddlePaddle/PaddleNLP/releases/tag/v2.4.0) Released!** - * 💎 NLP Tools: Released **[Pipelines](./pipelines)** which supports turn-key construction of search engine and question answering systems. It features a flexible design that is applicable for all kinds of NLP systems so you can build end-to-end NLP pipelines like Legos! +## News - * 🔨 Industrial application: Release **[Complete Solution of Text Classification](./applications/text_classification)** covering various scenarios of text classification: multi-class, multi-label and hierarchical, it also supports **few-shot learning** and the training and optimization of **TrustAI**. Upgrade for [**UIE**](./model_zoo/uie) and release **UIE-M**, support both Chinese and English information extraction in a single model; release the data distillation solution for UIE to break the bottleneck of time-consuming of inference. +* **2023.1.12: Release of PaddleNLP v2.5]()** - * 🍭 AIGC: Release code generation SOTA model [**CodeGen**](./examples/code_generation/codegen) that supports multiple programming languages code generation. Integrate [**Text to Image Model**](https://github.com/PaddlePaddle/PaddleNLP/blob/develop/docs/model_zoo/taskflow.md#%E6%96%87%E5%9B%BE%E7%94%9F%E6%88%90) DALL·E Mini, Disco Diffusion, Stable Diffusion, let's play and have some fun! + * 🔨 NLP Tools: [[PPDiffusers](./ppdiffusers), a domestically produced diffusion model toolbox, has been released. It integrates multiple diffusion model parameters and components, provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). + * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. + * 💪 Framework Upgrade: Pre-training model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. + * 🤝 Ecological Collaboration: 🤗Huggingface hub officially supports PaddleNLP pre-trained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to experience PaddleNLP pre-trained model effects on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). + +* **September 6, 2022: Release of PaddleNLP v2.4]()** + + * 🔨 NLP Tools: [NLP Pipeline System Pipelines](./pipelines) has been released, supporting the rapid construction of search engines and question-answering systems, and can be extended to support various NLP systems, making it easy, flexible, and efficient to solve NLP tasks like building blocks! + * 💎 Industrial Applications: A new [text classification full-process application solution](./applications/text_classification) has been added, covering various scenarios such as multi-classification, multi-label, and hierarchical classification, supporting small-sample learning and TrustAI trustworthy computing model training and tuning. + * 🍭 AIGC: The SOTA model [CodeGen](https://github.com/PaddlePaddle/PaddleNLP/blob/develop/examples/code_generation/codegen) for code generation in various programming languages has been added. + * 💪 Framework Upgrade: [Automatic Model Compression API](./docs/compression.md) has been released, which automatically cuts and quantizes models, greatly reducing the threshold for using model compression technology. [Small Sample Prompt](./applications/text_classification/multi_class/few-shot) capability has been released, integrating classic algorithms such as PET, P-Tuning, and RGL. - * 💪 Framework upgrade: Release [**Auto Model Compression API**](./docs/compression.md), supports for pruning and quantization automatically, lower the barriers of model compression; Release [**Few-shot Prompt**](./applications/text_classification/multi_class/few-shot), includes the algorithms such as PET, P-Tuning and RGL. From ffcf034a0e1e4a1f7b6e57624a8055bd05e1aea7 Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Wed, 3 May 2023 09:36:40 +0800 Subject: [PATCH 2/8] Update README_en.md --- README_en.md | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/README_en.md b/README_en.md index 914c33fb621f..85d10f70b1fe 100644 --- a/README_en.md +++ b/README_en.md @@ -1,26 +1,36 @@ [简体中文🀄](./README.MD) | **English🌎** -

+

-* * * * * +------------------------------------------------------------------------------------------ -

+

+ + + + + + + + + +

-

Features | Installation | Quick Start | API Documentation | Community +

Features | Installation | Quick Start | API Reference | Community **PaddleNLP** is a natural language processing development library that is both **easy to use** and **powerful**. It aggregates high-quality pre-trained models in the industry and provides a **plug-and-play** development experience, covering a model library for various NLP scenarios. With practical examples from industry practices, PaddleNLP can meet the needs of developers who require **flexible customization**. -## News +## News 📢 -* **2023.1.12: Release of PaddleNLP v2.5]()** +* **2023.1.12: [Release of PaddleNLP v2.5]()** - * 🔨 NLP Tools: [[PPDiffusers](./ppdiffusers), a domestically produced diffusion model toolbox, has been released. It integrates multiple diffusion model parameters and components, provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). + * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), a domestically produced diffusion model toolbox, has been released. It integrates multiple diffusion model parameters and components, provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. * 💪 Framework Upgrade: Pre-training model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. * 🤝 Ecological Collaboration: 🤗Huggingface hub officially supports PaddleNLP pre-trained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to experience PaddleNLP pre-trained model effects on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). -* **September 6, 2022: Release of PaddleNLP v2.4]()** +* **September 6, 2022: [Release of PaddleNLP v2.4]()** * 🔨 NLP Tools: [NLP Pipeline System Pipelines](./pipelines) has been released, supporting the rapid construction of search engines and question-answering systems, and can be extended to support various NLP systems, making it easy, flexible, and efficient to solve NLP tasks like building blocks! * 💎 Industrial Applications: A new [text classification full-process application solution](./applications/text_classification) has been added, covering various scenarios such as multi-classification, multi-label, and hierarchical classification, supporting small-sample learning and TrustAI trustworthy computing model training and tuning. From a4f56d73ebe2c7d166863f038fa72272cc8d2334 Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 14:57:12 +0800 Subject: [PATCH 3/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index 85d10f70b1fe..9b6aebbdad6a 100644 --- a/README_en.md +++ b/README_en.md @@ -25,7 +25,7 @@ * **2023.1.12: [Release of PaddleNLP v2.5]()** - * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), a domestically produced diffusion model toolbox, has been released. It integrates multiple diffusion model parameters and components, provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). + * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), our cross-modal diffusion model toolbox based on PaddlePaddle, has been released! It provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. * 💪 Framework Upgrade: Pre-training model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. * 🤝 Ecological Collaboration: 🤗Huggingface hub officially supports PaddleNLP pre-trained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to experience PaddleNLP pre-trained model effects on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). From 999073bdc18b26d055583ec811c1a7c75423934c Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 14:57:52 +0800 Subject: [PATCH 4/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index 9b6aebbdad6a..68f1dc732d6d 100644 --- a/README_en.md +++ b/README_en.md @@ -27,7 +27,7 @@ * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), our cross-modal diffusion model toolbox based on PaddlePaddle, has been released! It provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. - * 💪 Framework Upgrade: Pre-training model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. + * 💪 Framework Upgrade: Pretrained model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. * 🤝 Ecological Collaboration: 🤗Huggingface hub officially supports PaddleNLP pre-trained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to experience PaddleNLP pre-trained model effects on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). * **September 6, 2022: [Release of PaddleNLP v2.4]()** From d7277353229e3a2fb4fb84138e78ffe78381031e Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 14:59:43 +0800 Subject: [PATCH 5/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index 68f1dc732d6d..b279105e6ee4 100644 --- a/README_en.md +++ b/README_en.md @@ -28,7 +28,7 @@ * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), our cross-modal diffusion model toolbox based on PaddlePaddle, has been released! It provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. * 💪 Framework Upgrade: Pretrained model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. - * 🤝 Ecological Collaboration: 🤗Huggingface hub officially supports PaddleNLP pre-trained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to experience PaddleNLP pre-trained model effects on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). + * 🤝 Community: 🤗Huggingface hub officially supports PaddleNLP pretrained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to try out PaddleNLP pretrained models on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle). * **September 6, 2022: [Release of PaddleNLP v2.4]()** From 993bd0eb2ee11e310f880f8a4b08df35076fc970 Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 15:00:07 +0800 Subject: [PATCH 6/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index b279105e6ee4..3d7d4f13cb3d 100644 --- a/README_en.md +++ b/README_en.md @@ -35,7 +35,7 @@ * 🔨 NLP Tools: [NLP Pipeline System Pipelines](./pipelines) has been released, supporting the rapid construction of search engines and question-answering systems, and can be extended to support various NLP systems, making it easy, flexible, and efficient to solve NLP tasks like building blocks! * 💎 Industrial Applications: A new [text classification full-process application solution](./applications/text_classification) has been added, covering various scenarios such as multi-classification, multi-label, and hierarchical classification, supporting small-sample learning and TrustAI trustworthy computing model training and tuning. * 🍭 AIGC: The SOTA model [CodeGen](https://github.com/PaddlePaddle/PaddleNLP/blob/develop/examples/code_generation/codegen) for code generation in various programming languages has been added. - * 💪 Framework Upgrade: [Automatic Model Compression API](./docs/compression.md) has been released, which automatically cuts and quantizes models, greatly reducing the threshold for using model compression technology. [Small Sample Prompt](./applications/text_classification/multi_class/few-shot) capability has been released, integrating classic algorithms such as PET, P-Tuning, and RGL. + * 💪 Framework Upgrade: [Automatic Model Compression API](./docs/compression.md) has been released, which automatically cuts and quantizes models, greatly reducing the threshold for using model compression technology. [Few-shot Prompt](./applications/text_classification/multi_class/few-shot) capability has been released, integrating classic algorithms such as PET, P-Tuning, and RGL. From ccc534f272ffceed18c1ac78e5ea242c886cd862 Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 15:00:28 +0800 Subject: [PATCH 7/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index 3d7d4f13cb3d..9340405a771b 100644 --- a/README_en.md +++ b/README_en.md @@ -19,7 +19,7 @@

Features | Installation | Quick Start | API Reference | Community -**PaddleNLP** is a natural language processing development library that is both **easy to use** and **powerful**. It aggregates high-quality pre-trained models in the industry and provides a **plug-and-play** development experience, covering a model library for various NLP scenarios. With practical examples from industry practices, PaddleNLP can meet the needs of developers who require **flexible customization**. +**PaddleNLP** is a NLP library that is both **easy to use** and **powerful**. It aggregates high-quality pretrained models in the industry and provides a **plug-and-play** development experience, covering a model library for various NLP scenarios. With practical examples from industry practices, PaddleNLP can meet the needs of developers who require **flexible customization**. ## News 📢 From 7f5157654646847a3f85bfa99965326c388ae69d Mon Sep 17 00:00:00 2001 From: Sal Saromines <82336164+Joker1718@users.noreply.github.com> Date: Thu, 4 May 2023 15:00:49 +0800 Subject: [PATCH 8/8] Update README_en.md Co-authored-by: Sijun He --- README_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README_en.md b/README_en.md index 9340405a771b..0fddc83ca8d7 100644 --- a/README_en.md +++ b/README_en.md @@ -26,7 +26,7 @@ * **2023.1.12: [Release of PaddleNLP v2.5]()** * 🔨 NLP Tools: [PPDiffusers](./ppdiffusers), our cross-modal diffusion model toolbox based on PaddlePaddle, has been released! It provides a complete training process for diffusion models, and supports FastDeploy inference acceleration and multi-hardware deployment (supports Ascend chips and Kunlun core deployment). - * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pre-trained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pre-trained model deployment. + * 💎 Industrial Applications: Information extraction, text classification, sentiment analysis, and intelligent question answering have all been newly upgraded. New releases include document information extraction [UIE-X](./applications/information_extraction/document), unified text classification [UTC](./applications/zero_shot_text_classification), unified sentiment analysis [UIE-Senta](./applications/sentiment_analysis/unified_sentiment_extraction) , and [unsupervised QA application](./applications/question_answering/unsupervised_qa). At the same time, the [ERNIE 3.0 Tiny v2](./model_zoo/ernie-tiny) series of pretrained small models have been released, which are more effective with low-resource and foreign data. They provide open-source end-to-end deployment solutions such as model pruning, model quantization, FastDeploy inference acceleration, and edge-side deployment to reduce the difficulty of pretrained model deployment. * 💪 Framework Upgrade: Pretrained model [parameter configuration unification](./paddlenlp/transformers/configuration_utils.py), saving and loading custom parameter configurations no longer requires additional development; [Trainer API](./docs/trainer.md) has added BF16 training, recompute recalculations, sharding, and other distributed capabilities. Large-scale pre-training model training can easily be accomplished through simple configuration. [Model Compression API](./docs/compression.md) supports quantization training, vocabulary compression, and other functions. The compressed model has smaller accuracy loss, and the memory consumption of model deployment is greatly reduced. [Data Augmentation API](./docs/dataaug.md) has been comprehensively upgraded to support three granularities of data augmentation strategy: character, word, and sentence, making it easy to customize data augmentation strategies. * 🤝 Community: 🤗Huggingface hub officially supports PaddleNLP pretrained models, supporting PaddleNLP Model and Tokenizer downloads and uploads directly from the 🤗Huggingface hub. Everyone is welcome to try out PaddleNLP pretrained models on the 🤗Huggingface hub [here](https://huggingface.co/PaddlePaddle).