update docs (#2498)

* update docs

* update

* update readme

* update readme

.github/workflows/deploy_docs.yml

@@ -3,6 +3,7 @@ on:
   push:
     branches: #设置更新哪个分支会更新站点
       - develop
+      - release/3.0-beta2
 permissions:
   contents: write
 jobs:

README_en.md

@@ -30,6 +30,8 @@ PaddleX 3.0 is a low-code development tool for AI models built on the PaddlePadd
 |                                                              [**PP-ChatOCRv3-doc**](https://paddlepaddle.github.io/PaddleX/latest/en/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.html)                                                              |                                                            [**Time Series Forecasting**](https://paddlepaddle.github.io/PaddleX/latest/en/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.html)                                                            |                                                              [**Time Series Anomaly Detection**](https://paddlepaddle.github.io/PaddleX/latest/en/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.html)                                                              |                                                         [**Time Series Classification**](https://paddlepaddle.github.io/PaddleX/latest/en/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.html)                                                         |
 | <img src="https://github.com/PaddlePaddle/PaddleX/assets/142379845/e3d97f4e-ab46-411c-8155-494c61492b0a" height="126px" width="180px"> | <img src="https://github.com/PaddlePaddle/PaddleX/assets/142379845/6e897bf6-35fe-45e6-a040-e9a1a20cfdf2" height="126px" width="180px"> | <img src="https://github.com/PaddlePaddle/PaddleX/assets/142379845/c54c66cc-da4f-4631-877b-43b0fbb192a6" height="126px" width="180px"> | <img src="https://github.com/PaddlePaddle/PaddleX/assets/142379845/0ce925b2-3776-4dde-8ce0-5156d5a2476e" height="126px" width="180px"> |
 
+
+
 ## 🌟 Why PaddleX ?
 
   🎨 **Rich Models One-click Call**: Integrate over **200 PaddlePaddle models** covering multiple key areas such as OCR, object detection, and time series forecasting into **19 pipelines**. Experience the model effects quickly through easy Python API calls. Also supports **more than 20 modules** for easy model combination use by developers.

docs/index.en.md

@@ -98,6 +98,7 @@ PaddleX 3.0 is a low-code development tool for AI models built on the PaddlePadd
     </table>
 
 
+
 ## 🌟 Why PaddleX ?
 
   🎨 <b>Rich Models One-click Call</b>: Integrate over <b>200 PaddlePaddle models</b> covering multiple key areas such as OCR, object detection, and time series forecasting into <b>19 pipelines</b>. Experience the model effects quickly through easy Python API calls. Also supports <b>more than 20 modules</b> for easy model combination use by developers.

docs/index.md

@@ -122,8 +122,7 @@ PaddleX 3.0 是基于飞桨框架构建的低代码开发工具，它集成了
 
  <b>PaddleX 致力于实现产线级别的模型训练、推理与部署。模型产线是指一系列预定义好的、针对特定AI任务的开发流程，其中包含能够独立完成某类任务的单模型（单功能模块）组合。</b>
 
-
- ## 📊 能力支持
+## 📊 能力支持
 
 PaddleX的各个产线均支持本地<b>快速推理</b>，部分模型支持在[AI Studio星河社区](https://aistudio.baidu.com/overview)上进行<b>在线体验</b>，您可以快速体验各个产线的预训练模型效果，如果您对产线的预训练模型效果满意，可以直接对产线进行[高性能推理](https://paddlepaddle.github.io/PaddleX/latest/pipeline_deploy/high_performance_inference.html)/[服务化部署](https://paddlepaddle.github.io/PaddleX/latest/pipeline_deploy/service_deploy.html)/[端侧部署](https://paddlepaddle.github.io/PaddleX/latest/pipeline_deploy/edge_deploy.html)，如果不满意，您也可以使用产线的<b>二次开发</b>能力，提升效果。完整的产线开发流程请参考[PaddleX产线使用概览](https://paddlepaddle.github.io/PaddleX/latest/pipeline_usage/pipeline_develop_guide.html)或各产线使用[教程](#-文档)。
 
@@ -345,8 +344,8 @@ PaddleX的各个产线均支持本地<b>快速推理</b>，部分模型支持在
 
 </table>
 
-
-> ❗注：以上功能均基于 GPU/CPU 实现。PaddleX 还可在昆仑芯、昇腾、寒武纪和海光等主流硬件上进行快速推理和二次开发。下表详细列出了模型产线的支持情况，具体支持的模型列表请参阅[模型列表(昆仑芯XPU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_xpu.html)/[模型列表(昇腾NPU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_npu.html)/[模型列表(寒武纪MLU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_mlu.html)/[模型列表(海光DCU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_dcu.html)。我们正在适配更多的模型，并在主流硬件上推动高性能和服务化部署的实施。
+!!! tip
+     以上功能均基于 GPU/CPU 实现。PaddleX 还可在昆仑芯、昇腾、寒武纪和海光等主流硬件上进行快速推理和二次开发。下表详细列出了模型产线的支持情况，具体支持的模型列表请参阅[模型列表(昆仑芯XPU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_xpu.html)/[模型列表(昇腾NPU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_npu.html)/[模型列表(寒武纪MLU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_mlu.html)/[模型列表(海光DCU)](https://paddlepaddle.github.io/PaddleX/latest/support_list/model_list_dcu.html)。我们正在适配更多的模型，并在主流硬件上推动高性能和服务化部署的实施。
 
 🔥🔥 <b>国产化硬件能力支持</b>
 

docs/module_usage/tutorials/cv_modules/instance_segmentation.md

@@ -11,7 +11,7 @@ comments: true
 
 <table>
 <tr>
-<th>模型</th><th>Model Download Link</th>
+<th>模型</th><th>模型下载链接</th>
 <th>Mask AP</th>
 <th>GPU推理耗时（ms）</th>
 <th>CPU推理耗时 (ms)</th>
@@ -19,7 +19,7 @@ comments: true
 <th>介绍</th>
 </tr>
 <tr>
-<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">训练模型</a></td>
 <td>50.6</td>
 <td>132.693</td>
 <td>4896.17</td>
@@ -27,7 +27,7 @@ comments: true
 <td rowspan="5">Mask-RT-DETR 是一种基于RT-DETR的实例分割模型，通过采用最优性能的更好的PP-HGNetV2作为骨干网络，构建了MaskHybridEncoder编码器，引入了IOU-aware Query Selection 技术，使其在相同推理耗时上取得了SOTA实例分割精度。</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">训练模型</a></td>
 <td>45.7</td>
 <td>46.5059</td>
 <td>2575.92</td>
@@ -41,7 +41,7 @@ comments: true
 
 <table>
 <tr>
-<th>模型</th><th>Model Download Link</th>
+<th>模型</th><th>模型下载链接</th>
 <th>Mask AP</th>
 <th>GPU推理耗时（ms）</th>
 <th>CPU推理耗时 (ms)</th>
@@ -49,7 +49,7 @@ comments: true
 <th>介绍</th>
 </tr>
 <tr>
-<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">训练模型</a></td>
 <td>36.3</td>
 <td>-</td>
 <td>-</td>
@@ -57,14 +57,14 @@ comments: true
 <td rowspan="2">Cascade-MaskRCNN 是一种改进的Mask RCNN实例分割模型，通过级联多个检测器，利用不同IOU阈值优化分割结果，解决检测与推理阶段的mismatch问题，提高了实例分割的准确性。</td>
 </tr>
 <tr>
-<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">训练模型</a></td>
 <td>39.1</td>
 <td>-</td>
 <td>-</td>
 <td>254.7 M</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">训练模型</a></td>
 <td>50.6</td>
 <td>132.693</td>
 <td>4896.17</td>
@@ -72,35 +72,35 @@ comments: true
 <td rowspan="5">Mask-RT-DETR 是一种基于RT-DETR的实例分割模型，通过采用最优性能的更好的PP-HGNetV2作为骨干网络，构建了MaskHybridEncoder编码器，引入了IOU-aware Query Selection 技术，使其在相同推理耗时上取得了SOTA实例分割精度。</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">训练模型</a></td>
 <td>45.7</td>
 <td>46.5059</td>
 <td>2575.92</td>
 <td>113.6 M</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-M</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-M_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-M_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-M</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-M_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-M_pretrained.pdparams">训练模型</a></td>
 <td>42.7</td>
 <td>36.8329</td>
 <td>-</td>
 <td>66.6 M</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-S_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-S_pretrained.pdparams">训练模型</a></td>
 <td>41.0</td>
 <td>33.5007</td>
 <td>-</td>
 <td>51.8 M</td>
 </tr>
 <tr>
-<td>Mask-RT-DETR-X</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-X_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-X_pretrained.pdparams">Trained Model</a></td>
+<td>Mask-RT-DETR-X</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-X_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-X_pretrained.pdparams">训练模型</a></td>
 <td>47.5</td>
 <td>75.755</td>
 <td>3358.04</td>
 <td>237.5 M</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-FPN_pretrained.pdparams">训练模型</a></td>
 <td>35.6</td>
 <td>-</td>
 <td>-</td>
@@ -108,35 +108,35 @@ comments: true
 <td rowspan="6">Mask R-CNN是由华盛顿首例即现投影卡的一个全任务深度学习模型，能够在一个模型中完成图片实例的分类和定位，并结合图像级的遮罩（Mask）来完成分割任务。</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNet50-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-vd-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-vd-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNet50-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-vd-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-vd-FPN_pretrained.pdparams">训练模型</a></td>
 <td>36.4</td>
 <td>-</td>
 <td>-</td>
 <td>157.5 M</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNet50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNet50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50_pretrained.pdparams">训练模型</a></td>
 <td>32.8</td>
 <td>-</td>
 <td>-</td>
 <td>128.7 M</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNet101-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNet101-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-FPN_pretrained.pdparams">训练模型</a></td>
 <td>36.6</td>
 <td>-</td>
 <td>-</td>
 <td>225.4 M</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNet101-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-vd-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-vd-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNet101-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-vd-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-vd-FPN_pretrained.pdparams">训练模型</a></td>
 <td>38.1</td>
 <td>-</td>
 <td>-</td>
 <td>225.1 M</td>
 </tr>
 <tr>
-<td>MaskRCNN-ResNeXt101-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNeXt101-vd-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNeXt101-vd-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>MaskRCNN-ResNeXt101-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNeXt101-vd-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNeXt101-vd-FPN_pretrained.pdparams">训练模型</a></td>
 <td>39.5</td>
 <td>-</td>
 <td>-</td>
@@ -144,7 +144,7 @@ comments: true
 <td></td>
 </tr>
 <tr>
-<td>PP-YOLOE_seg-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_seg-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_seg-S_pretrained.pdparams">Trained Model</a></td>
+<td>PP-YOLOE_seg-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_seg-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_seg-S_pretrained.pdparams">训练模型</a></td>
 <td>32.5</td>
 <td>-</td>
 <td>-</td>
@@ -152,7 +152,7 @@ comments: true
 <td>PP-YOLOE_seg 是一种基于PP-YOLOE的实例分割模型。该模型沿用了PP-YOLOE的backbone和head，通过设计PP-YOLOE实例分割头，大幅提升了实例分割的性能和推理速度。</td>
 </tr>
 <tr>
-<td>SOLOv2</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SOLOv2_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SOLOv2_pretrained.pdparams">Trained Model</a></td>
+<td>SOLOv2</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SOLOv2_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SOLOv2_pretrained.pdparams">训练模型</a></td>
 <td>35.5</td>
 <td>-</td>
 <td>-</td>

docs/module_usage/tutorials/cv_modules/object_detection.md

@@ -66,6 +66,7 @@ comments: true
 </table>
 
 > ❗ 以上列出的是目标检测模块重点支持的<b>6个核心模型</b>，该模块总共支持<b>37个模型</b>，完整的模型列表如下：
+
 <details><summary> 👉模型列表详情</summary>
 
 <table>

docs/module_usage/tutorials/cv_modules/semantic_segmentation.md

@@ -36,6 +36,7 @@ comments: true
 </tr>
 </tbody>
 </table>
+
 > ❗ 以上列出的是语义分割模块重点支持的<b>2个核心模型</b>，该模块总共支持<b>18个模型</b>，完整的模型列表如下：
 
 <details><summary> 👉模型列表详情</summary>

docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md

@@ -13,7 +13,6 @@ The face recognition pipeline is an end-to-end system dedicated to solving face
 
 <b>The face recognition pipeline includes a face detection module and a face feature module</b>, with several models in each module. Which models to use can be selected based on the benchmark data below. <b>If you prioritize model accuracy, choose models with higher accuracy; if you prioritize inference speed, choose models with faster inference; if you prioritize model size, choose models with smaller storage requirements</b>.
 
-<details><summary> 👉Model List Details</summary>
 
 <p><b>Face Detection Module</b>:</p>
 <table>
@@ -97,7 +96,7 @@ The face recognition pipeline is an end-to-end system dedicated to solving face
 </tr>
 </tbody>
 </table>
-<p>Note: The above accuracy metrics are Accuracy scores measured on the AgeDB-30, CFP-FP, and LFW datasets, respectively. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</p></details>
+<p>Note: The above accuracy metrics are Accuracy scores measured on the AgeDB-30, CFP-FP, and LFW datasets, respectively. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</p>
 
 ## 2. Quick Start
 The pre-trained model pipelines provided by PaddleX can be quickly experienced. You can experience the effects of the face recognition pipeline online or locally using command-line or Python.

docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.md

@@ -13,8 +13,6 @@ comments: true
 
 <b>人脸识别产线中包含了人脸检测模块和人脸特征模块</b>，每个模块中包含了若干模型，具体使用哪些模型，您可以根据下边的 benchmark 数据来选择。<b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
-
 <p><b>人脸检测模块：</b></p>
 <table>
 <thead>
@@ -98,7 +96,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p>注：以上精度指标是分别在 AgeDB-30、CFP-FP 和 LFW 数据集上测得的 Accuracy。所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</p></details>
+<p>注：以上精度指标是分别在 AgeDB-30、CFP-FP 和 LFW 数据集上测得的 Accuracy。所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验人脸识别产线的效果，也可以在本地使用命令行或 Python 体验人脸识别产线的效果。

docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md

@@ -14,7 +14,6 @@ PP-ShiTuV2 is a practical general image recognition system mainly composed of th
 
 <b>The General Image Recognition Pipeline includes the mainbody detection module and the image feature module</b>, with several models to choose. You can select the model to use based on the benchmark data below. <b>If you prioritize model precision, choose a model with higher precision. If you prioritize inference speed, choose a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage size</b>.
 
-<summary> 👉Model List Details</summary>
 
 <b>Object Detection Module:</b>
 

docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md

@@ -14,7 +14,6 @@ PP-ShiTuV2 是一个实用的通用图像识别系统，主要由主体检测、
 
 <b>通用图像识别产线中包含了主体检测模块和图像特征模块</b>，有若干模型可供选择，您可以根据下边的 benchmark 数据来选择使用的模型。<b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<summary> 👉模型列表详情</summary>
 
 <b>主体检测模块：</b>
 

docs/pipeline_usage/tutorials/cv_pipelines/image_classification.en.md

@@ -11,6 +11,81 @@ Image classification is a technique that assigns images to predefined categories
 
 <b>The General Image Classification Pipeline includes an image classification module. If you prioritize model accuracy, choose a model with higher accuracy. If you prioritize inference speed, select a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage size.</b>
 
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Top1 Acc(%)</th>
+<th>GPU Inference Time (ms)</th>
+<th>CPU Inference Time (ms)</th>
+<th>Model Storage Size (M)</th>
+</tr>
+<tr>
+<td>CLIP_vit_base_patch16_224</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/CLIP_vit_base_patch16_224_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/CLIP_vit_base_patch16_224_pretrained.pdparams">Trained Model</a></td>
+<td>85.36</td>
+<td>13.1957</td>
+<td>285.493</td>
+<td>306.5 M</td>
+</tr>
+<tr>
+<td>MobileNetV3_small_x1_0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MobileNetV3_small_x1_0_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MobileNetV3_small_x1_0_pretrained.pdparams">Trained Model</a></td>
+<td>68.2</td>
+<td>6.00993</td>
+<td>12.9598</td>
+<td>10.5 M</td>
+</tr>
+<tr>
+<td>PP-HGNet_small</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNet_small_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNet_small_pretrained.pdparams">Trained Model</a></td>
+<td>81.51</td>
+<td>5.50661</td>
+<td>119.041</td>
+<td>86.5 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B0_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B0_pretrained.pdparams">Trained Model</a></td>
+<td>77.77</td>
+<td>6.53694</td>
+<td>23.352</td>
+<td>21.4 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B4</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B4_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B4_pretrained.pdparams">Trained Model</a></td>
+<td>83.57</td>
+<td>9.66407</td>
+<td>54.2462</td>
+<td>70.4 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B6</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B6_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B6_pretrained.pdparams">Trained Model</a></td>
+<td>86.30</td>
+<td>21.226</td>
+<td>255.279</td>
+<td>268.4 M</td>
+</tr>
+<tr>
+<td>PP-LCNet_x1_0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LCNet_x1_0_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LCNet_x1_0_pretrained.pdparams">Trained Model</a></td>
+<td>71.32</td>
+<td>3.84845</td>
+<td>9.23735</td>
+<td>10.5 M</td>
+</tr>
+<tr>
+<td>ResNet50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/ResNet50_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ResNet50_pretrained.pdparams">Trained Model</a></td>
+<td>76.5</td>
+<td>9.62383</td>
+<td>64.8135</td>
+<td>90.8 M</td>
+</tr>
+<tr>
+<td>SwinTransformer_tiny_patch4_window7_224</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SwinTransformer_tiny_patch4_window7_224_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams">Trained Model</a></td>
+<td>81.10</td>
+<td>8.54846</td>
+<td>156.306</td>
+<td>100.1 M</td>
+</tr>
+</table>
+
+> ❗ The above list features the <b>9 core models</b> that the image classification module primarily supports. In total, this module supports <b>80 models</b>. The complete list of models is as follows:
+
 <details><summary> 👉Details of Model List</summary>
 
 <table>
@@ -531,7 +606,7 @@ Image classification is a technique that assigns images to predefined categories
 </tr>
 <tr>
 <tr>
-<td>StarNet-S1</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/StarNet-S1_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/StarNet-S1_pretrained.pdparams">Trained Model</a></td><td>
+<td>StarNet-S1</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/StarNet-S1_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/StarNet-S1_pretrained.pdparams">Trained Model</a></td>
 <td>73.6</td>
 <td>9.895</td>
 <td>23.0465</td>

docs/pipeline_usage/tutorials/cv_pipelines/image_classification.md

@@ -11,14 +11,87 @@ comments: true
 
 <b>通用图像分类产线中包含了图像分类模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>Top1 Acc(%)</th>
+<th>GPU推理耗时 (ms)</th>
+<th>CPU推理耗时 (ms)</th>
+<th>模型存储大小 (M)</th>
+</tr>
+<tr>
+<td>CLIP_vit_base_patch16_224</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/CLIP_vit_base_patch16_224_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/CLIP_vit_base_patch16_224_pretrained.pdparams">训练模型</a></td>
+<td>85.36</td>
+<td>13.1957</td>
+<td>285.493</td>
+<td>306.5 M</td>
+</tr>
+<tr>
+<td>MobileNetV3_small_x1_0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MobileNetV3_small_x1_0_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MobileNetV3_small_x1_0_pretrained.pdparams">训练模型</a></td>
+<td>68.2</td>
+<td>6.00993</td>
+<td>12.9598</td>
+<td>10.5 M</td>
+</tr>
+<tr>
+<td>PP-HGNet_small</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNet_small_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNet_small_pretrained.pdparams">训练模型</a></td>
+<td>81.51</td>
+<td>5.50661</td>
+<td>119.041</td>
+<td>86.5 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B0_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B0_pretrained.pdparams">训练模型</a></td>
+<td>77.77</td>
+<td>6.53694</td>
+<td>23.352</td>
+<td>21.4 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B4</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B4_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B4_pretrained.pdparams">训练模型</a></td>
+<td>83.57</td>
+<td>9.66407</td>
+<td>54.2462</td>
+<td>70.4 M</td>
+</tr>
+<tr>
+<td>PP-HGNetV2-B6</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-HGNetV2-B6_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-HGNetV2-B6_pretrained.pdparams">训练模型</a></td>
+<td>86.30</td>
+<td>21.226</td>
+<td>255.279</td>
+<td>268.4 M</td>
+</tr>
+<tr>
+<td>PP-LCNet_x1_0</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LCNet_x1_0_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LCNet_x1_0_pretrained.pdparams">训练模型</a></td>
+<td>71.32</td>
+<td>3.84845</td>
+<td>9.23735</td>
+<td>10.5 M</td>
+</tr>
+<tr>
+<td>ResNet50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/ResNet50_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ResNet50_pretrained.pdparams">训练模型</a></td>
+<td>76.5</td>
+<td>9.62383</td>
+<td>64.8135</td>
+<td>90.8 M</td>
+</tr>
+<tr>
+<td>SwinTransformer_tiny_patch4_window7_224</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SwinTransformer_tiny_patch4_window7_224_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams">训练模型</a></td>
+<td>81.10</td>
+<td>8.54846</td>
+<td>156.306</td>
+<td>100.1 M</td>
+</tr>
+</table>
 
+> ❗ 以上列出的是图像分类模块重点支持的<b>9个核心模型</b>，该模块总共支持<b>80个模型</b>，完整的模型列表如下：
+<details><summary> 👉模型列表详情</summary>
 <table>
 <tr>
 <th>模型</th><th>模型下载链接</th>
 <th>Top1 Acc(%)</th>
 <th>GPU推理耗时 (ms)</th>
-<th>CPU推理耗时（ms）</th>
+<th>CPU推理耗时 (ms)</th>
 <th>模型存储大小 (M)</th>
 <th>介绍</th>
 </tr>
@@ -428,7 +501,6 @@ comments: true
 <td>32.1 M</td>
 </tr>
 <tr>
-<tr>
 <td>PP-LCNetV2_base</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LCNetV2_base_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LCNetV2_base_pretrained.pdparams">训练模型</a></td>
 <td>77.05</td>
 <td>5.23428</td>
@@ -451,7 +523,6 @@ comments: true
 <td>14.6 M</td>
 </tr>
 <tr>
-<tr>
 <td>ResNet18_vd</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/ResNet18_vd_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ResNet18_vd_pretrained.pdparams">训练模型</a></td>
 <td>72.3</td>
 <td>3.53048</td>
@@ -530,7 +601,6 @@ comments: true
 <td>266.0 M</td>
 </tr>
 <tr>
-<tr>
 <td>StarNet-S1</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/StarNet-S1_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/StarNet-S1_pretrained.pdparams">训练模型</a></td>
 <td>73.6</td>
 <td>9.895</td>
@@ -560,7 +630,6 @@ comments: true
 <td>28.9 M</td>
 </tr>
 <tr>
-<tr>
 <td>SwinTransformer_base_patch4_window7_224</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SwinTransformer_base_patch4_window7_224_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SwinTransformer_base_patch4_window7_224_pretrained.pdparams">训练模型</a></td>
 <td>83.37</td>
 <td>16.9848</td>
@@ -603,8 +672,7 @@ comments: true
 <td>156.306</td>
 <td>100.1 M</td>
 </tr>
-<tr>
-</tr></tr></tr></tr></tr></table>
+</table>
 
 <p><b>注：以上精度指标为 <a href="https://www.image-net.org/index.php">ImageNet-1k</a> 验证集 Top1 Acc。所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
 

docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.en.md

@@ -11,8 +11,6 @@ Image multi-label classification is a technique that assigns multiple relevant c
 
 <b>The General Image Multi-Label Classification Pipeline includes a module for image multi-label classification. If you prioritize model accuracy, choose a model with higher accuracy. If you prioritize inference speed, choose a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage size.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <table>
 <thead>
 <tr>
@@ -54,7 +52,7 @@ Image multi-label classification is a technique that assigns multiple relevant c
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are mAP for the multi-label classification task on </b><a href="https://cocodataset.org/#home">COCO2017</a><b>. The GPU inference time for all models is based on an NVIDIA Tesla T4 machine with FP32 precision. The CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are mAP for the multi-label classification task on </b><a href="https://cocodataset.org/#home">COCO2017</a><b>. The GPU inference time for all models is based on an NVIDIA Tesla T4 machine with FP32 precision. The CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 PaddleX supports experiencing the effects of the General Image Multi-Label Classification Pipeline locally using command line or Python.

docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.md

@@ -11,10 +11,6 @@ comments: true
 
 <b>通用图像多标签分类产线中包含了图像多标签分类模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-
-
-<details><summary> 👉模型列表详情</summary>
-
 <table>
 <thead>
 <tr>
@@ -56,7 +52,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 </b><a href="https://cocodataset.org/#home">COCO2017</a><b> 的多标签分类任务mAP。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标为 </b><a href="https://cocodataset.org/#home">COCO2017</a><b> 的多标签分类任务mAP。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 
 ## 2. 快速开始

docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.en.md

@@ -12,32 +12,74 @@ Instance segmentation is a computer vision task that not only identifies the obj
 
 <b>The General Instance Segmentation Pipeline includes a</b> <b>Object Detection</b> <b>module. If you prioritize model precision, choose a model with higher precision. If you prioritize inference speed, choose a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage size.</b>
 
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Mask AP</th>
+<th>GPU Inference Time (ms)</th>
+<th>CPU Inference Time (ms)</th>
+<th>Model Size (M)</th>
+<th>Description</th>
+</tr>
+<tr>
+<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">Trained Model</a></td>
+<td>50.6</td>
+<td>132.693</td>
+<td>4896.17</td>
+<td>449.9 M</td>
+<td rowspan="5">Mask-RT-DETR is an instance segmentation model based on RT-DETR. By adopting the high-performance PP-HGNetV2 as the backbone network and constructing a MaskHybridEncoder encoder, along with introducing IOU-aware Query Selection technology, it achieves state-of-the-art (SOTA) instance segmentation accuracy with the same inference time.</td>
+</tr>
+<tr>
+<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">Trained Model</a></td>
+<td>45.7</td>
+<td>46.5059</td>
+<td>2575.92</td>
+<td>113.6 M</td>
+</tr>
+</table>
+
+> ❗ The above list features the <b>2 core models</b> that the image classification module primarily supports. In total, this module supports <b>15 models</b>. The complete list of models is as follows:
+
 <details><summary> 👉Model List Details</summary>
 
 <table>
-<thead>
 <tr>
-<th>Model Name</th><th>Model Download Link</th>
+<th>Model</th><th>Model Download Link</th>
 <th>Mask AP</th>
 <th>GPU Inference Time (ms)</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
+<th>Description</th>
+</tr>
+<tr>
+<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>36.3</td>
+<td>-</td>
+<td>-</td>
+<td>254.8 M</td>
+<td rowspan="2">Cascade-MaskRCNN is an improved Mask RCNN instance segmentation model that utilizes multiple detectors in a cascade, optimizing segmentation results by leveraging different IOU thresholds to address the mismatch between detection and inference stages, thereby enhancing instance segmentation accuracy.</td>
+</tr>
+<tr>
+<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">Trained Model</a></td>
+<td>39.1</td>
+<td>-</td>
+<td>-</td>
+<td>254.7 M</td>
 </tr>
-</thead>
-<tbody>
 <tr>
 <td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">Trained Model</a></td>
 <td>50.6</td>
 <td>132.693</td>
 <td>4896.17</td>
-<td>449.9</td>
+<td>449.9 M</td>
+<td rowspan="5">Mask-RT-DETR is an instance segmentation model based on RT-DETR. By adopting the high-performance PP-HGNetV2 as the backbone network and constructing a MaskHybridEncoder encoder, along with introducing IOU-aware Query Selection technology, it achieves state-of-the-art (SOTA) instance segmentation accuracy with the same inference time.</td>
 </tr>
 <tr>
 <td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">Trained Model</a></td>
 <td>45.7</td>
 <td>46.5059</td>
 <td>2575.92</td>
-<td>113.6</td>
+<td>113.6 M</td>
 </tr>
 <tr>
 <td>Mask-RT-DETR-M</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-M_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-M_pretrained.pdparams">Trained Model</a></td>
@@ -61,25 +103,12 @@ Instance segmentation is a computer vision task that not only identifies the obj
 <td>237.5 M</td>
 </tr>
 <tr>
-<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">Trained Model</a></td>
-<td>36.3</td>
-<td>-</td>
-<td>-</td>
-<td>254.8</td>
-</tr>
-<tr>
-<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">Trained Model</a></td>
-<td>39.1</td>
-<td>-</td>
-<td>-</td>
-<td>254.7</td>
-</tr>
-<tr>
 <td>MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-FPN_pretrained.pdparams">Trained Model</a></td>
 <td>35.6</td>
 <td>-</td>
 <td>-</td>
 <td>157.5 M</td>
+<td rowspan="6">Mask R-CNN is a full-task deep learning model from Facebook AI Research (FAIR) that can perform object classification and localization in a single model, combined with image-level masks to complete segmentation tasks.</td>
 </tr>
 <tr>
 <td>MaskRCNN-ResNet50-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-vd-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-vd-FPN_pretrained.pdparams">Trained Model</a></td>
@@ -93,7 +122,7 @@ Instance segmentation is a computer vision task that not only identifies the obj
 <td>32.8</td>
 <td>-</td>
 <td>-</td>
-<td>127.8 M</td>
+<td>128.7 M</td>
 </tr>
 <tr>
 <td>MaskRCNN-ResNet101-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-FPN_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-FPN_pretrained.pdparams">Trained Model</a></td>
@@ -115,6 +144,7 @@ Instance segmentation is a computer vision task that not only identifies the obj
 <td>-</td>
 <td>-</td>
 <td>370.0 M</td>
+<td></td>
 </tr>
 <tr>
 <td>PP-YOLOE_seg-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_seg-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_seg-S_pretrained.pdparams">Trained Model</a></td>
@@ -122,6 +152,7 @@ Instance segmentation is a computer vision task that not only identifies the obj
 <td>-</td>
 <td>-</td>
 <td>31.5 M</td>
+<td>PP-YOLOE_seg is an instance segmentation model based on PP-YOLOE. This model inherits PP-YOLOE's backbone and head, significantly enhancing instance segmentation performance and inference speed through the design of a PP-YOLOE instance segmentation head.</td>
 </tr>
 <tr>
 <td>SOLOv2</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SOLOv2_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SOLOv2_pretrained.pdparams">Trained Model</a></td>
@@ -129,10 +160,11 @@ Instance segmentation is a computer vision task that not only identifies the obj
 <td>-</td>
 <td>-</td>
 <td>179.1 M</td>
+<td> SOLOv2 is a real-time instance segmentation algorithm that segments objects by location. This model is an improved version of SOLO, achieving a good balance between accuracy and speed through the introduction of mask learning and mask NMS.</td>
 </tr>
-</tbody>
 </table>
-<p><b>Note: The above accuracy metrics are Mask AP(0.5:0.95) on the </b><a href="https://cocodataset.org/#home">COCO2017</a><b> validation set. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+
+<p><b>Note: The above accuracy metrics are based on the Mask AP of the <a href="https://cocodataset.org/#home">COCO2017</a> validation set. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
 
 ## 2. Quick Start
 The pre-trained model pipelines provided by PaddleX allow for quick experience of the effects. You can experience the effects of the General Instance Segmentation Pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.md

@@ -9,34 +9,76 @@ comments: true
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/instance_segmentation/01.png">
 
-<b>通用实例分割产线中包含了</b><b>目标检测</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
+<b>通用实例分割产线中包含了</b><b>实例分割</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
+
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>Mask AP</th>
+<th>GPU推理耗时（ms）</th>
+<th>CPU推理耗时 (ms)</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">训练模型</a></td>
+<td>50.6</td>
+<td>132.693</td>
+<td>4896.17</td>
+<td>449.9 M</td>
+<td rowspan="5">Mask-RT-DETR 是一种基于RT-DETR的实例分割模型，通过采用最优性能的更好的PP-HGNetV2作为骨干网络，构建了MaskHybridEncoder编码器，引入了IOU-aware Query Selection 技术，使其在相同推理耗时上取得了SOTA实例分割精度。</td>
+</tr>
+<tr>
+<td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">训练模型</a></td>
+<td>45.7</td>
+<td>46.5059</td>
+<td>2575.92</td>
+<td>113.6 M</td>
+</tr>
+</table>
+
+> ❗ 以上列出的是实例分割模块重点支持的<b>2个核心模型</b>，该模块总共支持<b>15个模型</b>，完整的模型列表如下：
 
 <details><summary> 👉模型列表详情</summary>
 
 <table>
-<thead>
 <tr>
 <th>模型</th><th>模型下载链接</th>
 <th>Mask AP</th>
 <th>GPU推理耗时（ms）</th>
-<th>CPU推理耗时（ms）</th>
-<th>模型存储大小（M)</th>
+<th>CPU推理耗时 (ms)</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">训练模型</a></td>
+<td>36.3</td>
+<td>-</td>
+<td>-</td>
+<td>254.8 M</td>
+<td rowspan="2">Cascade-MaskRCNN 是一种改进的Mask RCNN实例分割模型，通过级联多个检测器，利用不同IOU阈值优化分割结果，解决检测与推理阶段的mismatch问题，提高了实例分割的准确性。</td>
+</tr>
+<tr>
+<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">训练模型</a></td>
+<td>39.1</td>
+<td>-</td>
+<td>-</td>
+<td>254.7 M</td>
 </tr>
-</thead>
-<tbody>
 <tr>
 <td>Mask-RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-H_pretrained.pdparams">训练模型</a></td>
 <td>50.6</td>
 <td>132.693</td>
 <td>4896.17</td>
-<td>449.9</td>
+<td>449.9 M</td>
+<td rowspan="5">Mask-RT-DETR 是一种基于RT-DETR的实例分割模型，通过采用最优性能的更好的PP-HGNetV2作为骨干网络，构建了MaskHybridEncoder编码器，引入了IOU-aware Query Selection 技术，使其在相同推理耗时上取得了SOTA实例分割精度。</td>
 </tr>
 <tr>
 <td>Mask-RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-L_pretrained.pdparams">训练模型</a></td>
 <td>45.7</td>
 <td>46.5059</td>
 <td>2575.92</td>
-<td>113.6</td>
+<td>113.6 M</td>
 </tr>
 <tr>
 <td>Mask-RT-DETR-M</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Mask-RT-DETR-M_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Mask-RT-DETR-M_pretrained.pdparams">训练模型</a></td>
@@ -60,25 +102,12 @@ comments: true
 <td>237.5 M</td>
 </tr>
 <tr>
-<td>Cascade-MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-FPN_pretrained.pdparams">训练模型</a></td>
-<td>36.3</td>
-<td>-</td>
-<td>-</td>
-<td>254.8</td>
-</tr>
-<tr>
-<td>Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN_pretrained.pdparams">训练模型</a></td>
-<td>39.1</td>
-<td>-</td>
-<td>-</td>
-<td>254.7</td>
-</tr>
-<tr>
 <td>MaskRCNN-ResNet50-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-FPN_pretrained.pdparams">训练模型</a></td>
 <td>35.6</td>
 <td>-</td>
 <td>-</td>
 <td>157.5 M</td>
+<td rowspan="6">Mask R-CNN是由华盛顿首例即现投影卡的一个全任务深度学习模型，能够在一个模型中完成图片实例的分类和定位，并结合图像级的遮罩（Mask）来完成分割任务。</td>
 </tr>
 <tr>
 <td>MaskRCNN-ResNet50-vd-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet50-vd-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet50-vd-FPN_pretrained.pdparams">训练模型</a></td>
@@ -92,7 +121,7 @@ comments: true
 <td>32.8</td>
 <td>-</td>
 <td>-</td>
-<td>127.8 M</td>
+<td>128.7 M</td>
 </tr>
 <tr>
 <td>MaskRCNN-ResNet101-FPN</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/MaskRCNN-ResNet101-FPN_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/MaskRCNN-ResNet101-FPN_pretrained.pdparams">训练模型</a></td>
@@ -114,6 +143,7 @@ comments: true
 <td>-</td>
 <td>-</td>
 <td>370.0 M</td>
+<td></td>
 </tr>
 <tr>
 <td>PP-YOLOE_seg-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_seg-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_seg-S_pretrained.pdparams">训练模型</a></td>
@@ -121,6 +151,7 @@ comments: true
 <td>-</td>
 <td>-</td>
 <td>31.5 M</td>
+<td>PP-YOLOE_seg 是一种基于PP-YOLOE的实例分割模型。该模型沿用了PP-YOLOE的backbone和head，通过设计PP-YOLOE实例分割头，大幅提升了实例分割的性能和推理速度。</td>
 </tr>
 <tr>
 <td>SOLOv2</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SOLOv2_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SOLOv2_pretrained.pdparams">训练模型</a></td>
@@ -128,10 +159,11 @@ comments: true
 <td>-</td>
 <td>-</td>
 <td>179.1 M</td>
+<td> SOLOv2 是一种按位置分割物体的实时实例分割算法。该模型是SOLO的改进版本，通过引入掩码学习和掩码NMS，实现了精度和速度上取得良好平衡。</td>
 </tr>
-</tbody>
 </table>
-<p><b>注：以上精度指标为 </b><a href="https://cocodataset.org/#home">COCO2017</a><b> 验证集 Mask AP(0.5:0.95)。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+
+<p><b>注：以上精度指标为<a href="https://cocodataset.org/#home">COCO2017</a>验证集 Mask AP。所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用实例分割产线的效果，也可以在本地使用命令行或 Python 体验通用实例分割产线的效果。

docs/pipeline_usage/tutorials/cv_pipelines/object_detection.en.md

@@ -11,6 +11,64 @@ Object detection aims to identify the categories and locations of multiple objec
 
 
 
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>mAP(%)</th>
+<th>GPU Inference Time (ms)</th>
+<th>CPU Inference Time (ms)</th>
+<th>Model Storage Size (M)</th>
+<th>Description</th>
+</tr>
+<tr>
+<td>PicoDet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PicoDet-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PicoDet-L_pretrained.pdparams">Trained Model</a></td>
+<td>42.6</td>
+<td>16.6715</td>
+<td>169.904</td>
+<td>20.9 M</td>
+<td rowspan="2">PP-PicoDet is a lightweight object detection algorithm for full-size, wide-angle targets, considering the computational capacity of mobile devices. Compared to traditional object detection algorithms, PP-PicoDet has a smaller model size and lower computational complexity, achieving higher speed and lower latency while maintaining detection accuracy.</td>
+</tr>
+<tr>
+<td>PicoDet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PicoDet-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PicoDet-S_pretrained.pdparams">Trained Model</a></td>
+<td>29.1</td>
+<td>14.097</td>
+<td>37.6563</td>
+<td>4.4 M</td>
+</tr>
+<tr>
+<td>PP-YOLOE_plus-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-L_pretrained.pdparams">Trained Model</a></td>
+<td>52.9</td>
+<td>33.5644</td>
+<td>814.825</td>
+<td>185.3 M</td>
+<td rowspan="2">PP-YOLOE_plus is an upgraded version of the high-precision cloud-edge integrated model PP-YOLOE, developed by Baidu's PaddlePaddle vision team. By using the large-scale Objects365 dataset and optimizing preprocessing, it significantly enhances the model's end-to-end inference speed.</td>
+</tr>
+<tr>
+<td>PP-YOLOE_plus-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-S_pretrained.pdparams">Trained Model</a></td>
+<td>43.7</td>
+<td>16.8884</td>
+<td>223.059</td>
+<td>28.3 M</td>
+</tr>
+<tr>
+<td>RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-H_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-H_pretrained.pdparams">Trained Model</a></td>
+<td>56.3</td>
+<td>114.814</td>
+<td>3933.39</td>
+<td>435.8 M</td>
+<td rowspan="2">RT-DETR is the first real-time end-to-end object detector. The model features an efficient hybrid encoder to meet both model performance and throughput requirements, efficiently handling multi-scale features, and proposes an accelerated and optimized query selection mechanism to optimize the dynamics of decoder queries. RT-DETR supports flexible end-to-end inference speeds by using different decoders.</td>
+</tr>
+<tr>
+<td>RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-L_pretrained.pdparams">Trained Model</a></td>
+<td>53.0</td>
+<td>34.5252</td>
+<td>1454.27</td>
+<td>113.7 M</td>
+</tr>
+</table>
+
+> ❗ The above list features the <b>6 core models</b> that the image classification module primarily supports. In total, this module supports <b>37 models</b>. The complete list of models is as follows:
+
 <details><summary> 👉Details of Model List</summary>
 
 <table>

docs/pipeline_usage/tutorials/cv_pipelines/object_detection.md

@@ -11,6 +11,64 @@ comments: true
 
 <b>通用</b><b>目标检测</b><b>产线中包含了</b><b>目标检测</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>mAP(%)</th>
+<th>GPU推理耗时 (ms)</th>
+<th>CPU推理耗时 (ms)</th>
+<th>模型存储大小 (M)</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>PicoDet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PicoDet-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PicoDet-L_pretrained.pdparams">训练模型</a></td>
+<td>42.6</td>
+<td>16.6715</td>
+<td>169.904</td>
+<td>20.9 M</td>
+<td rowspan="2">PP-PicoDet是一种全尺寸、棱视宽目标的轻量级目标检测算法，它考虑移动端设备运算量。与传统目标检测算法相比，PP-PicoDet具有更小的模型尺寸和更低的计算复杂度，并在保证检测精度的同时更高的速度和更低的延迟。</td>
+</tr>
+<tr>
+<td>PicoDet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PicoDet-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PicoDet-S_pretrained.pdparams">训练模型</a></td>
+<td>29.1</td>
+<td>14.097</td>
+<td>37.6563</td>
+<td>4.4 M</td>
+</tr>
+<tr>
+<td>PP-YOLOE_plus-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-L_pretrained.pdparams">训练模型</a></td>
+<td>52.9</td>
+<td>33.5644</td>
+<td>814.825</td>
+<td>185.3 M</td>
+<td rowspan="2">PP-YOLOE_plus 是一种是百度飞桨视觉团队自研的云边一体高精度模型PP-YOLOE迭代优化升级的版本，通过使用Objects365大规模数据集、优化预处理，大幅提升了模型端到端推理速度。</td>
+</tr>
+<tr>
+<td>PP-YOLOE_plus-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-YOLOE_plus-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-YOLOE_plus-S_pretrained.pdparams">训练模型</a></td>
+<td>43.7</td>
+<td>16.8884</td>
+<td>223.059</td>
+<td>28.3 M</td>
+</tr>
+<tr>
+<td>RT-DETR-H</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-H_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-H_pretrained.pdparams">训练模型</a></td>
+<td>56.3</td>
+<td>114.814</td>
+<td>3933.39</td>
+<td>435.8 M</td>
+<td rowspan="2">RT-DETR是第一个实时端到端目标检测器。该模型设计了一个高效的混合编码器，满足模型效果与吞吐率的双需求，高效处理多尺度特征，并提出了加速和优化的查询选择机制，以优化解码器查询的动态化。RT-DETR支持通过使用不同的解码器来实现灵活端到端推理速度。</td>
+</tr>
+<tr>
+<td>RT-DETR-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/RT-DETR-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/RT-DETR-L_pretrained.pdparams">训练模型</a></td>
+<td>53.0</td>
+<td>34.5252</td>
+<td>1454.27</td>
+<td>113.7 M</td>
+</tr>
+</table>
+
+> ❗ 以上列出的是目标检测模块重点支持的<b>6个核心模型</b>，该模块总共支持<b>37个模型</b>，完整的模型列表如下：
+
 <details><summary> 👉模型列表详情</summary>
 
 <table>
@@ -295,8 +353,6 @@ comments: true
 
 <p><b>注：以上精度指标为<a href="https://cocodataset.org/#home">COCO2017</a>验证集 mAP(0.5:0.95)。所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
 
-</details>
-
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用目标检测产线的效果，也可以在本地使用命令行或 Python 体验通用目标检测产线的效果。
 

docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.en.md

@@ -11,7 +11,6 @@ Pedestrian attribute recognition is a key function in computer vision systems, u
 
 <b>The pedestrian attribute recognition pipeline includes a pedestrian detection module and a pedestrian attribute recognition module</b>, with several models in each module. Which models to use specifically can be selected based on the benchmark data below. <b>If you prioritize model accuracy, choose models with higher accuracy; if you prioritize inference speed, choose models with faster inference; if you prioritize model storage size, choose models with smaller storage</b>.
 
-<details><summary> 👉Model List Details</summary>
 
 <p><b>Pedestrian Detection Module</b>:</p>
 <table>
@@ -67,7 +66,7 @@ Pedestrian attribute recognition is a key function in computer vision systems, u
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are mA on PaddleX's internally built dataset. GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are mA on PaddleX's internally built dataset. GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 The pre-trained model pipelines provided by PaddleX can quickly demonstrate their effectiveness. You can experience the pedestrian attribute recognition pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.md

@@ -11,7 +11,6 @@ comments: true
 
 <b>行人属性识别产线中包含了行人检测模块和行人属性识别模块</b>，每个模块中包含了若干模型，具体使用哪些模型，您可以根据下边的 benchmark 数据来选择。<b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <p><b>行人检测模块：</b></p>
 <table>
@@ -67,7 +66,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 PaddleX 内部自建数据集 mA。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标为 PaddleX 内部自建数据集 mA。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验行人属性识别产线的效果，也可以在本地使用命令行或 Python 体验行人属性识别产线的效果。

docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md

@@ -9,8 +9,36 @@ Semantic segmentation is a computer vision technique that aims to assign each pi
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/semantic_segmentation/01.png">
 
-<details><summary> 👉 Model List Details</summary>
+<table>
+<thead>
+<tr>
+<th>Model Name</th><th>Model Download Link</th>
+<th>mIoU (%)</th>
+<th>GPU Inference Time (ms)</th>
+<th>CPU Inference Time (ms)</th>
+<th>Model Size (M)</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>OCRNet_HRNet-W48</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W48_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W48_pretrained.pdparams">Trained Model</a></td>
+<td>82.15</td>
+<td>78.9976</td>
+<td>2226.95</td>
+<td>249.8 M</td>
+</tr>
+<tr>
+<td>PP-LiteSeg-T</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-T_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-T_pretrained.pdparams">Trained Model</a></td>
+<td>73.10</td>
+<td>7.6827</td>
+<td>138.683</td>
+<td>28.5 M</td>
+</tr>
+</tbody>
+</table>
+> ❗ The above list features the <b>2 core models</b> that the image classification module primarily supports. In total, this module supports <b>18 models</b>. The complete list of models is as follows:
 
+<details><summary> 👉Model List Details</summary>
 <table>
 <thead>
 <tr>
@@ -129,7 +157,7 @@ Semantic segmentation is a computer vision technique that aims to assign each pi
 <th>Model Name</th><th>Model Download Link</th>
 <th>mIoU (%)</th>
 <th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time (ms)</th>
+<th>CPU Inference Time</th>
 <th>Model Size (M)</th>
 </tr>
 </thead>

docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.md

@@ -11,112 +11,142 @@ comments: true
 
 <b>通用</b><b>语义分割</b><b>产线中包含了</b><b>语义分割</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
+<table>
+<thead>
+<tr>
+<th>模型名称</th><th>Model Download Link</th>
+<th>mloU（%）</th>
+<th>GPU推理耗时（ms）</th>
+<th>CPU推理耗时 (ms)</th>
+<th>模型存储大小（M)</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>OCRNet_HRNet-W48</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W48_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W48_pretrained.pdparams">Trained Model</a></td>
+<td>82.15</td>
+<td>78.9976</td>
+<td>2226.95</td>
+<td>249.8 M</td>
+</tr>
+<tr>
+<td>PP-LiteSeg-T</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-T_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-T_pretrained.pdparams">Trained Model</a></td>
+<td>73.10</td>
+<td>7.6827</td>
+<td>138.683</td>
+<td>28.5 M</td>
+</tr>
+</tbody>
+</table>
+
+> ❗ 以上列出的是语义分割模块重点支持的<b>2个核心模型</b>，该模块总共支持<b>18个模型</b>，完整的模型列表如下：
+
 <details><summary> 👉模型列表详情</summary>
 
 <table>
 <thead>
 <tr>
-<th>模型</th><th>模型下载链接</th>
+<th>模型名称</th><th>Model Download Link</th>
 <th>mloU（%）</th>
 <th>GPU推理耗时（ms）</th>
-<th>CPU推理耗时（ms）</th>
+<th>CPU推理耗时 (ms)</th>
 <th>模型存储大小（M)</th>
 </tr>
 </thead>
 <tbody>
 <tr>
-<td>Deeplabv3_Plus-R50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3_Plus-R50_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3_Plus-R50_pretrained.pdparams">训练模型</a></td>
+<td>Deeplabv3_Plus-R50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3_Plus-R50_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3_Plus-R50_pretrained.pdparams">Trained Model</a></td>
 <td>80.36</td>
 <td>61.0531</td>
 <td>1513.58</td>
 <td>94.9 M</td>
 </tr>
 <tr>
-<td>Deeplabv3_Plus-R101</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3_Plus-R101_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3_Plus-R101_pretrained.pdparams">训练模型</a></td>
+<td>Deeplabv3_Plus-R101</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3_Plus-R101_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3_Plus-R101_pretrained.pdparams">Trained Model</a></td>
 <td>81.10</td>
 <td>100.026</td>
 <td>2460.71</td>
 <td>162.5 M</td>
 </tr>
 <tr>
-<td>Deeplabv3-R50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3-R50_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3-R50_pretrained.pdparams">训练模型</a></td>
+<td>Deeplabv3-R50</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3-R50_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3-R50_pretrained.pdparams">Trained Model</a></td>
 <td>79.90</td>
 <td>82.2631</td>
 <td>1735.83</td>
 <td>138.3 M</td>
 </tr>
 <tr>
-<td>Deeplabv3-R101</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3-R101_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3-R101_pretrained.pdparams">训练模型</a></td>
+<td>Deeplabv3-R101</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/Deeplabv3-R101_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/Deeplabv3-R101_pretrained.pdparams">Trained Model</a></td>
 <td>80.85</td>
 <td>121.492</td>
 <td>2685.51</td>
 <td>205.9 M</td>
 </tr>
 <tr>
-<td>OCRNet_HRNet-W18</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W18_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W18_pretrained.pdparams">训练模型</a></td>
+<td>OCRNet_HRNet-W18</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W18_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W18_pretrained.pdparams">Trained Model</a></td>
 <td>80.67</td>
 <td>48.2335</td>
 <td>906.385</td>
 <td>43.1 M</td>
 </tr>
 <tr>
-<td>OCRNet_HRNet-W48</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W48_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W48_pretrained.pdparams">训练模型</a></td>
+<td>OCRNet_HRNet-W48</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/OCRNet_HRNet-W48_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/OCRNet_HRNet-W48_pretrained.pdparams">Trained Model</a></td>
 <td>82.15</td>
 <td>78.9976</td>
 <td>2226.95</td>
 <td>249.8 M</td>
 </tr>
 <tr>
-<td>PP-LiteSeg-T</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-T_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-T_pretrained.pdparams">训练模型</a></td>
+<td>PP-LiteSeg-T</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-T_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-T_pretrained.pdparams">Trained Model</a></td>
 <td>73.10</td>
 <td>7.6827</td>
 <td>138.683</td>
 <td>28.5 M</td>
 </tr>
 <tr>
-<td>PP-LiteSeg-B</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-B_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-B_pretrained.pdparams">训练模型</a></td>
+<td>PP-LiteSeg-B</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-LiteSeg-B_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LiteSeg-B_pretrained.pdparams">Trained Model</a></td>
 <td>75.25</td>
 <td>10.9935</td>
 <td>194.727</td>
 <td>47.0 M</td>
 </tr>
 <tr>
-<td>SegFormer-B0 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B0 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B0 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B0 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B0 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B0 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>76.73</td>
 <td>11.1946</td>
 <td>268.929</td>
 <td>13.2 M</td>
 </tr>
 <tr>
-<td>SegFormer-B1 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B1 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B1 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B1 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B1 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B1 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>78.35</td>
 <td>17.9998</td>
 <td>403.393</td>
 <td>48.5 M</td>
 </tr>
 <tr>
-<td>SegFormer-B2 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B2 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B2 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B2 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B2 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B2 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>81.60</td>
 <td>48.0371</td>
 <td>1248.52</td>
 <td>96.9 M</td>
 </tr>
 <tr>
-<td>SegFormer-B3 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B3 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B3 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B3 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B3 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B3 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>82.47</td>
 <td>64.341</td>
 <td>1666.35</td>
 <td>167.3 M</td>
 </tr>
 <tr>
-<td>SegFormer-B4 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B4 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B4 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B4 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B4 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B4 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>82.38</td>
 <td>82.4336</td>
 <td>1995.42</td>
 <td>226.7 M</td>
 </tr>
 <tr>
-<td>SegFormer-B5 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B5 (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B5 (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SegFormer-B5 (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SegFormer-B5 (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SegFormer-B5 (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>82.58</td>
 <td>97.3717</td>
 <td>2420.19</td>
@@ -124,41 +154,41 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 </b><a href="https://www.cityscapes-dataset.com/">Cityscapes</a><b> 数据集 mloU。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
+<p><b>以上模型精度指标测量自<a href="https://www.cityscapes-dataset.com/">Cityscapes</a>数据集。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p>
 <table>
 <thead>
 <tr>
-<th>模型</th><th>模型下载链接</th>
+<th>模型名称</th><th>Model Download Link</th>
 <th>mloU（%）</th>
 <th>GPU推理耗时（ms）</th>
-<th>CPU推理耗时（ms）</th>
+<th>CPU推理耗时</th>
 <th>模型存储大小（M)</th>
 </tr>
 </thead>
 <tbody>
 <tr>
-<td>SeaFormer_base(slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_base(slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_base(slice)_pretrained.pdparams">训练模型</a></td>
+<td>SeaFormer_base(slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_base(slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_base(slice)_pretrained.pdparams">Trained Model</a></td>
 <td>40.92</td>
 <td>24.4073</td>
 <td>397.574</td>
 <td>30.8 M</td>
 </tr>
 <tr>
-<td>SeaFormer_large (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_large (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_large (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SeaFormer_large (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_large (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_large (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>43.66</td>
 <td>27.8123</td>
 <td>550.464</td>
 <td>49.8 M</td>
 </tr>
 <tr>
-<td>SeaFormer_small (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_small (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_small (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SeaFormer_small (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_small (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_small (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>38.73</td>
 <td>19.2295</td>
 <td>358.343</td>
 <td>14.3 M</td>
 </tr>
 <tr>
-<td>SeaFormer_tiny (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_tiny (slice)_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_tiny (slice)_pretrained.pdparams">训练模型</a></td>
+<td>SeaFormer_tiny (slice)</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/SeaFormer_tiny (slice)_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/SeaFormer_tiny (slice)_pretrained.pdparams">Trained Model</a></td>
 <td>34.58</td>
 <td>13.9496</td>
 <td>330.132</td>
@@ -166,7 +196,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 </b><a href="https://groups.csail.mit.edu/vision/datasets/ADE20K/">ADE20k</a><b> 数据集, slice 表示对输入图像进行了切图操作。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>SeaFormer系列模型的精度指标测量自<a href="https://groups.csail.mit.edu/vision/datasets/ADE20K/">ADE20k</a>数据集。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p></details>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用语义分割产线的效果，也可以在本地使用命令行或 Python 体验通用语义分割产线的效果。

docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md

@@ -11,8 +11,6 @@ Small object detection is a specialized technique for identifying tiny objects w
 
 <b>The small object detection pipeline includes a small object detection module. If you prioritize model accuracy, choose a model with higher accuracy. If you prioritize inference speed, select a model with faster inference. If you prioritize model storage size, opt for a model with a smaller storage size.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <table>
 <thead>
 <tr>
@@ -47,7 +45,7 @@ Small object detection is a specialized technique for identifying tiny objects w
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are based on the </b><a href="https://github.com/VisDrone/VisDrone-Dataset">VisDrone-DET</a><b> validation set mAP(0.5:0.95). All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are based on the </b><a href="https://github.com/VisDrone/VisDrone-Dataset">VisDrone-DET</a><b> validation set mAP(0.5:0.95). All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 PaddleX supports experiencing the small object detection pipeline's effects through command line or Python locally.

docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md

@@ -11,7 +11,6 @@ comments: true
 
 <b>小目标检测产线中包含了小目标检测模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <table>
 <thead>
@@ -47,7 +46,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 </b><a href="https://github.com/VisDrone/VisDrone-Dataset">VisDrone-DET</a><b> 验证集 mAP(0.5:0.95)。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标为 </b><a href="https://github.com/VisDrone/VisDrone-Dataset">VisDrone-DET</a><b> 验证集 mAP(0.5:0.95)。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 支持在本地使用命令行或 Python 体验小目标检测产线的效果。

docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.en.md

@@ -11,8 +11,6 @@ Vehicle attribute recognition is a crucial component in computer vision systems.
 
 <b>The vehicle attribute recognition pipeline includes a vehicle detection module and a vehicle attribute recognition module</b>, with several models in each module. Which models to use can be selected based on the benchmark data below. <b>If you prioritize model accuracy, choose models with higher accuracy; if you prioritize inference speed, choose models with faster inference; if you prioritize model storage size, choose models with smaller storage</b>.
 
-<details><summary> 👉Model List Details</summary>
-
 <p><b>Vehicle Detection Module</b>:</p>
 <table>
 <tr>
@@ -64,7 +62,7 @@ Vehicle attribute recognition is a crucial component in computer vision systems.
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are mA on the VeRi dataset. GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are mA on the VeRi dataset. GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 The pre-trained models provided by PaddleX can quickly demonstrate results. You can experience the effects of the vehicle attribute recognition pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.md

@@ -11,7 +11,6 @@ comments: true
 
 <b>车辆属性识别产线中包含了车辆检测模块和车辆属性识别模块</b>，每个模块中包含了若干模型，具体使用哪些模型，您可以根据下边的 benchmark 数据来选择。<b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <p><b>车辆检测模块：</b></p>
 <table>
@@ -64,7 +63,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标为 VeRi 数据集mA。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标为 VeRi 数据集mA。GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为 8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验车辆属性识别产线的效果，也可以在本地使用命令行或 Python 体验车辆属性识别产线的效果。

docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md

@@ -13,8 +13,6 @@ The General OCR Pipeline is designed to solve text recognition tasks, extracting
 
 <b>The General OCR Pipeline comprises a text detection module and a text recognition module</b>, each containing several models. The specific models to use can be selected based on the benchmark data provided below. <b>If you prioritize model accuracy, choose models with higher accuracy. If you prioritize inference speed, choose models with faster inference. If you prioritize model size, choose models with smaller storage requirements.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <p><b>Text detection module:</b></p>
 <table>
 <thead>
@@ -114,7 +112,7 @@ The General OCR Pipeline is designed to solve text recognition tasks, extracting
 </tr>
 </table>
 
-<p><b>Note: The evaluation set for the above accuracy metrics is the <a href="https://aistudio.baidu.com/competition/detail/1131/0/introduction">OCR End-to-End Recognition Task of the PaddleOCR Algorithm Model Challenge - Track 1</a> B-rank. GPU inference time for all models is based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The evaluation set for the above accuracy metrics is the <a href="https://aistudio.baidu.com/competition/detail/1131/0/introduction">OCR End-to-End Recognition Task of the PaddleOCR Algorithm Model Challenge - Track 1</a> B-rank. GPU inference time for all models is based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 PaddleX provides pre-trained models for the OCR Pipeline, allowing you to quickly experience its effects. You can try the General OCR Pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md

@@ -14,8 +14,6 @@ OCR（光学字符识别，Optical Character Recognition）是一种将图像中
 
 <b>通用OCR产线中包含了文本检测模块和文本识别模块</b>，每个模块中包含了若干模型，具体使用哪些模型，您可以根据下边的 benchmark 数据来选择。<b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
-
 <p><b>文本检测模块：</b></p>
 <table>
 <thead>
@@ -116,7 +114,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 </table>
 
-<p><b>注：以上精度指标的评估集是 <a href="https://aistudio.baidu.com/competition/detail/1131/0/introduction">PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务</a>B榜。 所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标的评估集是 <a href="https://aistudio.baidu.com/competition/detail/1131/0/introduction">PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务</a>B榜。 所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用OCR产线的效果，也可以在本地使用命令行或 Python 体验通用 OCR 产线的效果。

docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.en.md

@@ -14,7 +14,6 @@ Formula recognition is a technology that automatically identifies and extracts L
 
 <b>If you prioritize model accuracy, choose a model with higher accuracy. If you prioritize inference speed, select a model with faster inference. If you prioritize model size, choose a model with a smaller storage footprint.</b>
 
-<details><summary> 👉Model List Details</summary>
 
 <p><b>Layout Detection Module Models</b>:</p>
 <table>
@@ -63,7 +62,7 @@ Formula recognition is a technology that automatically identifies and extracts L
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are measured on the <a href="https://drive.google.com/drive/folders/13CA4vAmOmD_I_dSbvLp-Lf0s6KiaNfuO">LaTeX-OCR Formula Recognition Test Set</a>. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are measured on the <a href="https://drive.google.com/drive/folders/13CA4vAmOmD_I_dSbvLp-Lf0s6KiaNfuO">LaTeX-OCR Formula Recognition Test Set</a>. All GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 PaddleX supports experiencing the effects of the formula recognition pipeline through command line or Python locally.

docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md

@@ -14,7 +14,6 @@ comments: true
 
 <b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <p><b>版面区域检测模块模型：</b></p>
 <table>
@@ -63,7 +62,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标测量自 <a href="https://drive.google.com/drive/folders/13CA4vAmOmD_I_dSbvLp-Lf0s6KiaNfuO">LaTeX-OCR公式识别测试集</a>。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标测量自 <a href="https://drive.google.com/drive/folders/13CA4vAmOmD_I_dSbvLp-Lf0s6KiaNfuO">LaTeX-OCR公式识别测试集</a>。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 支持在本地使用命令行或 Python 体验公式识别产线的效果。

docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md

@@ -13,7 +13,6 @@ The <b>Seal Recognition</b> pipeline includes a layout area analysis module, a s
 
 <b>If you prioritize model accuracy, please choose a model with higher accuracy. If you prioritize inference speed, please choose a model with faster inference. If you prioritize model storage size, please choose a model with a smaller storage footprint.</b>
 
-<details><summary> 👉 Detailed Model List </summary>
 
 <p><b>Layout Analysis Module Models:</b></p>
 <table>
@@ -147,7 +146,7 @@ The <b>Seal Recognition</b> pipeline includes a layout area analysis module, a s
 </tr>
 </tbody>
 </table>
-<p><b>Note: The evaluation set for the above accuracy indicators is a self-built Chinese dataset from PaddleOCR, covering various scenarios such as street scenes, web images, documents, and handwriting. The text recognition subset includes 11,000 images. The GPU inference time for all models above is based on an NVIDIA Tesla T4 machine with a precision type of FP32. The CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads, and the precision type is also FP32.</b></p></details>
+<p><b>Note: The evaluation set for the above accuracy indicators is a self-built Chinese dataset from PaddleOCR, covering various scenarios such as street scenes, web images, documents, and handwriting. The text recognition subset includes 11,000 images. The GPU inference time for all models above is based on an NVIDIA Tesla T4 machine with a precision type of FP32. The CPU inference speed is based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads, and the precision type is also FP32.</b></p>
 
 ## 2.  Quick Start
 The pre trained model production line provided by PaddleX can quickly experience the effect. You can experience the effect of the seal recognition production line online, or use the command line or Python locally to experience the effect of the seal recognition production line.

docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md

@@ -15,7 +15,6 @@ comments: true
 
 <b>如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <p><b>版面区域分析模块模型：</b></p>
 <table>
@@ -152,7 +151,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标的评估集是 PaddleOCR 自建的中文数据集 ，覆盖街景、网图、文档、手写多个场景，其中文本识别包含 1.1w 张图片。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标的评估集是 PaddleOCR 自建的中文数据集 ，覆盖街景、网图、文档、手写多个场景，其中文本识别包含 1.1w 张图片。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在本地使用命令行或 Python 体验印章文本识别产线的效果。

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.en.md

@@ -11,8 +11,6 @@ Time series anomaly detection is a technique for identifying abnormal patterns o
 
 <b>The General Time Series Anomaly Detection Pipeline includes a time series anomaly detection module. If you prioritize model accuracy, choose a model with higher precision. If you prioritize inference speed, select a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage footprint.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <table>
 <thead>
 <tr>
@@ -61,7 +59,7 @@ Time series anomaly detection is a technique for identifying abnormal patterns o
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above precision metrics are measured on the </b><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/ts_anomaly_examples.tar">PSM</a><b> dataset. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above precision metrics are measured on the </b><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/ts_anomaly_examples.tar">PSM</a><b> dataset. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 The pre-trained model pipelines provided by PaddleX allow for quick experience of their effects. You can experience the effects of the General Time Series Anomaly Detection Pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.md

@@ -13,9 +13,6 @@ comments: true
 <b>通用</b><b>时序异常检测</b><b>产线中包含了</b><b>时序异常检测</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
 
-
-<details><summary> 👉模型列表详情</summary>
-
 <table>
 <thead>
 <tr>
@@ -64,7 +61,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标测量自 </b><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/ts_anomaly_examples.tar">PSM</a><b> 数据集。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标测量自 </b><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/ts_anomaly_examples.tar">PSM</a><b> 数据集。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 
 ## 2. 快速开始

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.en.md

@@ -11,8 +11,6 @@ Time series classification is a technique that categorizes time-series data into
 
 <b>The General Time Series Classification Pipeline includes a Time Series Classification module.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <table>
 <thead>
 <tr>
@@ -29,7 +27,7 @@ Time series classification is a technique that categorizes time-series data into
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are measured on the <a href="https://paddlets.bj.bcebos.com/classification/UWaveGestureLibrary_TEST.csv">UWaveGestureLibrary</a> dataset. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
+<p><b>Note: The above accuracy metrics are measured on the <a href="https://paddlets.bj.bcebos.com/classification/UWaveGestureLibrary_TEST.csv">UWaveGestureLibrary</a> dataset. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 
 ## 2. Quick Start
 PaddleX provides pre-trained model pipelines that can be quickly experienced. You can experience the effects of the General Time Series Classification Pipeline online or locally using command line or Python.

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.md

@@ -12,8 +12,6 @@ comments: true
 
 <b>通用</b><b>时序分类</b><b>产线中包含了</b><b>时序分类</b><b>模块</b>。
 
-<details><summary> 👉模型列表详情</summary>
-
 <table>
 <thead>
 <tr>
@@ -30,7 +28,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标测量自 <a href="https://paddlets.bj.bcebos.com/classification/UWaveGestureLibrary_TEST.csv">UWaveGestureLibrary</a> 数据集。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标测量自 <a href="https://paddlets.bj.bcebos.com/classification/UWaveGestureLibrary_TEST.csv">UWaveGestureLibrary</a> 数据集。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用时序分类产线的效果，也可以在本地使用命令行或 Python 体验通用时序分类产线的效果。

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.en.md

@@ -11,8 +11,6 @@ Time series forecasting is a technique that utilizes historical data to predict
 
 <b>The General Time Series Forecasting Pipeline includes a time series forecasting module. If you prioritize model accuracy, choose a model with higher accuracy. If you prioritize inference speed, select a model with faster inference. If you prioritize model storage size, choose a model with a smaller storage size.</b>
 
-<details><summary> 👉Model List Details</summary>
-
 <table>
 <thead>
 <tr>
@@ -67,8 +65,7 @@ Time series forecasting is a technique that utilizes historical data to predict
 </tr>
 </tbody>
 </table>
-<p><b>Note: The above accuracy metrics are measured on <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/Etth1.tar">ETTH1</a>. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p></details>
-
+<p><b>Note: The above accuracy metrics are measured on <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/Etth1.tar">ETTH1</a>. All model GPU inference times are based on an NVIDIA Tesla T4 machine with FP32 precision. CPU inference speeds are based on an Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz with 8 threads and FP32 precision.</b></p>
 ## 2. Quick Start
 The pre-trained model pipelines provided by PaddleX allow for quick experience of their effects. You can experience the effects of the General Time Series Forecasting Pipeline online or locally using command line or Python.
 

docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.md

@@ -11,7 +11,6 @@ comments: true
 
 <b>通用</b><b>时序预测</b><b>产线中包含了</b><b>时序预测</b><b>模块，如您更考虑模型精度，请选择精度较高的模型，如您更考虑模型推理速度，请选择推理速度较快的模型，如您更考虑模型存储大小，请选择存储大小较小的模型</b>。
 
-<details><summary> 👉模型列表详情</summary>
 
 <table>
 <thead>
@@ -67,7 +66,7 @@ comments: true
 </tr>
 </tbody>
 </table>
-<p><b>注：以上精度指标测量自 <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/Etth1.tar">ETTH1</a>。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p></details>
+<p><b>注：以上精度指标测量自 <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/data/Etth1.tar">ETTH1</a>。以上所有模型 GPU 推理耗时基于 NVIDIA Tesla T4 机器，精度类型为 FP32， CPU 推理速度基于 Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz，线程数为8，精度类型为 FP32。</b></p>
 
 ## 2. 快速开始
 PaddleX 所提供的预训练的模型产线均可以快速体验效果，你可以在线体验通用时序预测产线的效果，也可以在本地使用命令行或 Python 体验通用时序预测产线的效果。