Fix en doc (#3377)

* fix en doc

* fix table in formula doc

docs/module_usage/tutorials/ocr_modules/formula_recognition.en.md

@@ -14,30 +14,35 @@ The formula recognition module is a crucial component of OCR (Optical Character
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Avg-BLEU(%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <td>UniMERNet</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">Training Model</a></td>
 <td>86.13</td>
-<td>2266.96</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td>UniMERNet is a formula recognition model developed by Shanghai AI Lab. It uses Donut Swin as the encoder and MBartDecoder as the decoder. The model is trained on a dataset of one million samples, including simple formulas, complex formulas, scanned formulas, and handwritten formulas, significantly improving the recognition accuracy of real-world formulas.</td>
 <tr>
 <td>PP-FormulaNet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">Training Model</a></td>
 <td>87.12</td>
-<td>202.25</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td rowspan="2">PP-FormulaNet is an advanced formula recognition model developed by the Baidu PaddlePaddle Vision Team. The PP-FormulaNet-S version uses PP-HGNetV2-B4 as its backbone network. Through parallel masking and model distillation techniques, it significantly improves inference speed while maintaining high recognition accuracy, making it suitable for applications requiring fast inference. The PP-FormulaNet-L version, on the other hand, uses Vary_VIT_B as its backbone network and is trained on a large-scale formula dataset, showing significant improvements in recognizing complex formulas compared to PP-FormulaNet-S.</td>
 </tr>
 <td>PP-FormulaNet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">Training Model</a></td>
 <td>92.13</td>
-<td>1976.52</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <tr>
 <td>LaTeX_OCR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">Training Model</a></td>
 <td>71.63</td>
-<td>-</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td>LaTeX-OCR is a formula recognition algorithm based on an autoregressive large model. It uses Hybrid ViT as the backbone network and a transformer as the decoder, significantly improving the accuracy of formula recognition.</td>
 </tr>
@@ -126,6 +131,13 @@ The explanations for the methods, parameters, etc., are as follows:
 <td>None</td>
 <td>None</td>
 </tr>
+<tr>
+<td><code>use_hpip</code></td>
+<td>Whether to enable high-performance inference. </td>
+<td><code>bool</code></td>
+<td>None</td>
+<td><code>False</code></td>
+</tr>
 </table>
 
 * The `model_name` must be specified. After specifying `model_name`, the default model parameters built into PaddleX are used. If `model_dir` is specified, the user-defined model is used.
@@ -455,7 +467,7 @@ Other related parameters can be set by modifying the `Global` and `Evaluate` fie
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 
 <p>When evaluating the model, you need to specify the model weights file path. Each configuration file has a default weight save path built-in. If you need to change it, simply set it by appending a command line parameter, such as <code>-o Evaluate.weight_path=./output/best_accuracy/best_accuracy.pdparams</code>.</p>
-<p>After completing the model evaluation, an <code>evaluate_result.json</code> file will be produced, which records the evaluation results, specifically, whether the evaluation task was completed successfully and the model's evaluation metrics, including recall1、recall5、mAP；</p></details>
+<p>After completing the model evaluation, an <code>evaluate_result.json</code> file will be produced, which records the evaluation results, specifically, whether the evaluation task was completed successfully and the model's evaluation metrics, including exp_rate；</p></details>
 
 
 ### <b>4.4 Model Inference and Integration</b>
@@ -483,4 +495,4 @@ Other related parameters can be set by modifying the `Global` and `Predict` fiel
 
 The weights you produce can be directly integrated into the formula recognition module. Refer to the Python example code in [Quick Integration](#iii-quick-integration), and simply replace the model with the path to your trained model.
 
-You can also use the PaddleX high-performance inference plugin to optimize the inference process of your model and further improve efficiency. For detailed procedures, please refer to the [PaddleX High-Performance Inference Guide](../../../pipeline_deploy/high_performance_inference.en.md).
+You can also use the PaddleX high-performance inference plugin to optimize the inference process of your model and further improve efficiency. For detailed procedures, please refer to the [PaddleX High-Performance Inference Guide](../../../pipeline_deploy/high_performance_inference.en.md).

docs/module_usage/tutorials/ocr_modules/formula_recognition.md

@@ -13,31 +13,36 @@ comments: true
 <tr>
 <th>模型</th><th>模型下载链接</th>
 <th>Avg-BLEU(%)</th>
-<th>GPU推理耗时 (ms)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
 <th>模型存储大小 (M)</th>
 <th>介绍</th>
 </tr>
 <td>UniMERNet</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">训练模型</a></td>
 <td>86.13</td>
-<td>2266.96</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td>UniMERNet是由上海AI Lab研发的一款公式识别模型。该模型采用Donut Swin作为编码器，MBartDecoder作为解码器，并通过在包含简单公式、复杂公式、扫描捕捉公式和手写公式在内的一百万数据集上进行训练，大幅提升了模型对真实场景公式的识别准确率</td>
 <tr>
 <td>PP-FormulaNet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">训练模型</a></td>
 <td>87.12</td>
-<td>202.25</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td rowspan="2">PP-FormulaNet 是由百度飞桨视觉团队开发的一款先进的公式识别模型，支持5万个常见LateX源码词汇的识别。PP-FormulaNet-S 版本采用了 PP-HGNetV2-B4 作为其骨干网络，通过并行掩码和模型蒸馏等技术，大幅提升了模型的推理速度，同时保持了较高的识别精度，适用于简单印刷公式、跨行简单印刷公式等场景。而 PP-FormulaNet-L 版本则基于 Vary_VIT_B 作为骨干网络，并在大规模公式数据集上进行了深入训练，在复杂公式的识别方面，相较于PP-FormulaNet-S表现出显著的提升，适用于简单印刷公式、复杂印刷公式、手写公式等场景。 </td>
 
 </tr>
 <td>PP-FormulaNet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">训练模型</a></td>
 <td>92.13</td>
-<td>1976.52</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <tr>
 <td>LaTeX_OCR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">训练模型</a></td>
 <td>71.63</td>
-<td>-</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td>LaTeX-OCR是一种基于自回归大模型的公式识别算法，通过采用 Hybrid ViT 作为骨干网络，transformer作为解码器，显著提升了公式识别的准确性。</td>
 </tr>

docs/module_usage/tutorials/ocr_modules/text_recognition.en.md

@@ -20,35 +20,36 @@ The text recognition module is the core component of an OCR (Optical Character R
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>81.53</td>
 <td>6.65 / 6.65</td>
 <td>32.92 / 32.92</td>
-<td></td>
+<td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
-<td>78.20</td>
+<td>78.74</td>
 <td>4.82 / 4.82</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The PP-OCRv4 recognition model is further upgraded based on PP-OCRv3. Under comparable speed conditions, the effect in Chinese and English scenarios is further improved, and the average recognition accuracy of the 80-language multilingual model is increased by more than 8%.</td>
+<td>
+The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
-<td>79.20</td>
+<td>80.61 </td>
 <td>6.58 / 6.58</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>A high-precision server-side text recognition model, featuring high accuracy, fast speed, and multilingual support. It is suitable for text recognition tasks in various scenarios.</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>70.39</td>
 <td>4.81 / 4.81</td>
 <td>16.10 / 5.31</td>
-<td></td>
-<td>The ultra-lightweight English text recognition model released by PaddleOCR in May 2023. It is small in size and fast in speed, and can achieve millisecond-level prediction on CPU. Compared with the PP-OCRv3 English model, the recognition accuracy is improved by 6%, and it is suitable for text recognition tasks in various scenarios.</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
 </tr>
 </table>
 
@@ -68,35 +69,35 @@ The text recognition module is the core component of an OCR (Optical Character R
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>81.53</td>
 <td>6.65 / 6.65</td>
 <td>32.92 / 32.92</td>
-<td></td>
+<td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
-<td>78.20</td>
+<td>78.74</td>
 <td>4.82 / 4.82</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The PP-OCRv4 recognition model is an upgrade from PP-OCRv3. Under comparable speed conditions, the effect in Chinese and English scenarios is further improved. The average recognition accuracy of the 80 multilingual models is increased by more than 8%.</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
-<td>79.20</td>
+<td>80.61 </td>
 <td>6.58 / 6.58</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>A high-precision server text recognition model, featuring high accuracy, fast speed, and multilingual support. It is suitable for text recognition tasks in various scenarios.</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>72.96</td>
 <td>5.87 / 5.87</td>
 <td>9.07 / 4.28</td>
-<td></td>
-<td>An ultra-lightweight OCR model suitable for mobile applications. It adopts an encoder-decoder structure based on Transformer and enhances recognition accuracy and efficiency through techniques such as data augmentation and mixed precision training. The model size is 10.6M, making it suitable for deployment on resource-constrained devices. It can be used in scenarios such as mobile photo translation and business card recognition.</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
 </tr>
 </table>
 
@@ -112,8 +113,8 @@ The text recognition module is the core component of an OCR (Optical Character R
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.36801</td>
-<td>165.706</td>
+<td>8.08 / 8.08</td>
+<td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
 SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
@@ -133,8 +134,8 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>10.5047</td>
-<td>51.5647</td>
+<td>5.93 / 5.93</td>
+<td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
 </tr>
@@ -152,19 +153,19 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>[Latest] Further upgraded based on PP-OCRv3, with improved accuracy under comparable speed conditions.</td>
+<td> 70.39</td>
+<td>4.81 / 4.81</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Ultra-lightweight model, supporting English and numeric recognition.</td>
+<td>70.69</td>
+<td>5.44 / 5.44</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
 </tr>
 </table>
 
@@ -180,83 +181,84 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Korean Recognition</td>
+<td>60.21</td>
+<td>5.40 / 5.40</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Japanese Recognition</td>
+<td>45.69</td>
+<td>5.70 / 5.70</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Traditional Chinese Recognition</td>
+<td>82.06</td>
+<td>5.90 / 5.90</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Telugu Recognition</td>
+<td>95.88</td>
+<td>5.42 / 5.42</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Kannada Recognition</td>
+<td>96.96</td>
+<td>5.25 / 5.25</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Tamil Recognition</td>
+<td>76.83</td>
+<td>5.23 / 5.23</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Latin Recognition</td>
+<td>76.93</td>
+<td>5.20 / 5.20</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Arabic Script Recognition</td>
+<td>73.55</td>
+<td>5.35 / 5.35</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Cyrillic Script Recognition</td>
+<td>94.28</td>
+<td>5.23 / 5.23</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Devanagari Script Recognition</td>
+<td>96.44</td>
+<td>5.22 / 5.22</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
 
@@ -313,9 +315,9 @@ The visualized image is as follows:
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/refs/heads/main/images/modules/text_recog/general_ocr_rec_001.png"/>
 
-In the above Python script, the following steps are executed:
-* `create_model` instantiates the text recognition model (here, `PP-OCRv4_mobile_rec` is taken as an example)
-* The `predict` method of the text recognition model is called for inference prediction. The parameter of the `predict` method is `x`, which is used to input the data to be predicted. It supports multiple input types, and the specific instructions are as follows:
+The explanations for the methods, parameters, etc., are as follows:
+
+* The `create_model` instantiates the text recognition model (here, `PP-OCRv4_mobile_rec` is taken as an example), and the specific instructions are as follows:
 
 <table>
 <thead>
@@ -328,55 +330,65 @@ In the above Python script, the following steps are executed:
 </tr>
 </thead>
 <tr>
-<td><code>x</code></td>
-<td>Data to be predicted, supporting multiple input types</td>
-<td><code>Python Var</code>/<code>str</code>/<code>list</code></td>
-<td>
-<ul>
-<li><b>Python Variable</b>, such as image data represented by <code>numpy.ndarray</code></li>
-<li><b>File Path</b>, such as the local path of an image file: <code>/root/data/img.jpg</code></li>
-<li><b>URL Link</b>, such as the web URL of an image file: <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_rec_001.png">Example</a></li>
-<li><b>Local Directory</b>, the directory should contain the data files to be predicted, such as the local path: <code>/root/data/</code></li>
-<li><b>List</b>, the elements of the list should be of the above-mentioned data types, such as <code>[numpy.ndarray, numpy.ndarray]</code>, <code>[\"/root/data/img1.jpg\", \"/root/data/img2.jpg\"]</code>, <code>[\"/root/data1\", \"/root/data2\"]</code></li>
-</ul>
-</td>
+<td><code>model_name</code></td>
+<td>Name of the model</td>
+<td><code>str</code></td>
+<td>All model names supported by PaddleX</td>
 <td>None</td>
 </tr>
 <tr>
-<td><code>module_name</code></td>
-<td>Name of the single-function module</td>
+<td><code>model_dir</code></td>
+<td>Path to store the model</td>
 <td><code>str</code></td>
 <td>None</td>
-<td><code>text_recognition</code></td>
+<td>None</td>
 </tr>
 <tr>
-<td><code>model_name</code></td>
-<td>Name of the model</td>
-<td><code>str</code></td>
+<td><code>use_hpip</code></td>
+<td>Whether to enable high-performance inference. </td>
+<td><code>bool</code></td>
 <td>None</td>
-<td><code>PP-OCRv4_mobile_rec</code></td>
+<td><code>False</code></td>
 </tr>
+</table>
+
+* The `model_name` must be specified. After specifying `model_name`, the default model parameters built into PaddleX are used. If `model_dir` is specified, the user-defined model is used.
+
+
+* The `predict()` method of the formula recognition model is called for inference prediction. The `predict()` method has parameters `input` and `batch_size`, which are explained as follows:
+
+<table>
+<thead>
 <tr>
-<td><code>model_dir</code></td>
-<td>Path where the model is stored</td>
-<td><code>str</code></td>
+<th>Parameter</th>
+<th>Parameter Description</th>
+<th>Parameter Type</th>
+<th>Options</th>
+<th>Default Value</th>
+</tr>
+</thead>
+<tr>
+<td><code>input</code></td>
+<td>Data to be predicted, supporting multiple input types</td>
+<td><code>Python Var</code>/<code>str</code>/<code>list</code></td>
+<td>
+<ul>
+  <li><b>Python variable</b>, such as image data represented by <code>numpy.ndarray</code></li>
+  <li><b>File path</b>, such as the local path of an image file: <code>/root/data/img.jpg</code></li>
+  <li><b>URL link</b>, such as the network URL of an image file: <a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_formula_rec_001.png">Example</a></li>
+  <li><b>Local directory</b>, the directory should contain data files to be predicted, such as the local path: <code>/root/data/</code></li>
+  <li><b>List</b>, elements of the list must be of the above types of data, such as <code>[numpy.ndarray, numpy.ndarray]</code>, <code>["/root/data/img1.jpg", "/root/data/img2.jpg"]</code>, <code>["/root/data1", "/root/data2"]</code>, <code>[{"img": "/root/data1"}, {"img": "/root/data2/img.jpg"}]</code></li>
+</ul>
+</td>
 <td>None</td>
-<td><code>null</code></td>
 </tr>
 <tr>
 <td><code>batch_size</code></td>
 <td>Batch size</td>
 <td><code>int</code></td>
-<td>None</td>
+<td>Any integer</td>
 <td>1</td>
 </tr>
-<tr>
-<td><code>score_thresh</code></td>
-<td>Score threshold</td>
-<td><code>int</code></td>
-<td>None</td>
-<td><code>0</code></td>
-</tr>
 </table>
 
 * Process the prediction results. The prediction result for each sample is of `dict` type, and supports operations such as printing, saving as an image, and saving as a `json` file:
@@ -442,6 +454,27 @@ In the above Python script, the following steps are executed:
 </tr>
 </table>
 
+* Additionally, it supports obtaining the visualization image with results and the prediction results through attributes, as follows:
+
+<table>
+<thead>
+<tr>
+<th>Attribute</th>
+<th>Attribute Description</th>
+</tr>
+</thead>
+<tr>
+<td rowspan="1"><code>json</code></td>
+<td rowspan="1">Get the prediction result in <code>json</code> format</td>
+</tr>
+<tr>
+<td rowspan="1"><code>img</code></td>
+<td rowspan="1">Get the visualization image in <code>dict</code> format</td>
+</tr>
+</table>
+
+For more information on using PaddleX's single-model inference API, refer to the [PaddleX Single Model Python Script Usage Instructions](../../instructions/model_python_API.en.md).
+
 
 ## IV. Custom Development
 
@@ -596,12 +629,9 @@ Other related parameters can be set by modifying the `Global` and `Train` fields
 After completing model training, you can evaluate the specified model weights file on the validation set to verify the model's accuracy. Using PaddleX for model evaluation can be done with a single command:
 
 ```bash
-
-```bash
 python main.py -c paddlex/configs/modules/text_recognition/PP-OCRv4_mobile_rec.yaml \
     -o Global.mode=evaluate \
     -o Global.dataset_dir=./dataset/ocr_rec_dataset_examples
-
 ```
 Similar to model training, the following steps are required:
 
@@ -649,4 +679,4 @@ The text recognition module can be integrated into PaddleX pipelines such as the
 
 The weights you produce can be directly integrated into the text recognition module. Refer to the [Quick Integration](#iii-quick-integration) Python example code. Simply replace the model with the path to your trained model.
 
-You can also use the PaddleX high-performance inference plugin to optimize the inference process of your model and further improve efficiency. For detailed procedures, please refer to the [PaddleX High-Performance Inference Guide](../../../pipeline_deploy/high_performance_inference.en.md).
+You can also use the PaddleX high-performance inference plugin to optimize the inference process of your model and further improve efficiency. For detailed procedures, please refer to the [PaddleX High-Performance Inference Guide](../../../pipeline_deploy/high_performance_inference.en.md).

docs/module_usage/tutorials/video_modules/video_classification.en.md

@@ -119,7 +119,7 @@ The Python script above performs the following steps:
 </tr>
 <tr>
 <td><code> topk</code></td>
-<td>The top `topk` categories and corresponding classification probabilities of the prediction result；if not specified, the default configuration of the PaddleX official model will be used</td>
+<td>The top <code> topk</code> categories and corresponding classification probabilities of the prediction result；if not specified, the default configuration of the PaddleX official model will be used</td>
 <td><code>int</code></td>
 <td>None</td>
 <td><code>1</code></td>
@@ -162,7 +162,7 @@ The Python script above performs the following steps:
 </tr>
 <tr>
 <td><code>topk</code></td>
-<td>The top `topk` categories and corresponding classification probabilities of the prediction result</td>
+<td>The <code>topk</code> predicted classes and their corresponding probabilities; if not specified, the <code>topk</code> parameter specified in create_model will be used by default. If create_model also does not specify it, the default will be the PaddleX official model configuration.</td>
 <td><code>int</code></td>
 <td>None</td>
 <td><code>1</code></td>
@@ -232,7 +232,7 @@ The Python script above performs the following steps:
 </tr>
 </table>
 
-* Additionally, the prediction results can also be obtained through attributes, as follows:
+* Additionally, it also supports obtaining result visualization videos and <code>json</code> results through attributes, as follows:
 
 <table>
 <thead>

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

@@ -10,6 +10,7 @@ OCR (Optical Character Recognition) is a technology that converts text in images
 The General OCR pipeline is designed to solve text recognition tasks, extracting text information from images and outputting it in text form. This pipeline integrates the well-known end-to-end OCR series systems, PP-OCRv3 and PP-OCRv4, supporting recognition of over 80 languages. Additionally, it includes functions for image orientation correction and distortion correction. Based on this pipeline, precise text content prediction at the millisecond level on CPUs can be achieved, covering a wide range of applications including general, manufacturing, finance, and transportation sectors. The pipeline also provides flexible deployment options, supporting calls in various programming languages on multiple hardware platforms. Moreover, it offers the capability for secondary development, allowing you to train and optimize on your own dataset. The trained models can also be seamlessly integrated.
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/ocr/01.png"/>
+
 <b>The General OCR pipeline includes mandatory text detection and text recognition modules, as well as optional document image orientation classification, text image correction, and text line orientation classification modules.</b> The document image orientation classification and text image correction modules are integrated as a document preprocessing sub-line into the General OCR pipeline. Each module contains multiple models, and you can choose the model based on the benchmark test data below.
 
 <b>If you prioritize model accuracy, choose a high-accuracy model; if you prioritize inference speed, choose a faster inference model; if you care about model storage size, choose a smaller model.</b>
@@ -164,35 +165,35 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>81.53</td>
 <td>6.65 / 6.65</td>
 <td>32.92 / 32.92</td>
-<td></td>
+<td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
-<td>78.20</td>
+<td>78.74</td>
 <td>4.82 / 4.82</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The PP-OCRv4 recognition model is an upgrade from PP-OCRv3. Under comparable speed conditions, the effect in Chinese and English scenarios is further improved. The average recognition accuracy of the 80 multilingual models is increased by more than 8%.</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
-<td>79.20</td>
+<td>80.61 </td>
 <td>6.58 / 6.58</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>A high-precision server text recognition model, featuring high accuracy, fast speed, and multilingual support. It is suitable for text recognition tasks in various scenarios.</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
+<td>72.96</td>
 <td>5.87 / 5.87</td>
 <td>9.07 / 4.28</td>
-<td></td>
-<td>An ultra-lightweight OCR model suitable for mobile applications. It adopts an encoder-decoder structure based on Transformer and enhances recognition accuracy and efficiency through techniques such as data augmentation and mixed precision training. The model size is 10.6M, making it suitable for deployment on resource-constrained devices. It can be used in scenarios such as mobile photo translation and business card recognition.</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
 </tr>
 </table>
 
@@ -208,8 +209,8 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.36801</td>
-<td>165.706</td>
+<td>8.08 / 8.08</td>
+<td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
 SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
@@ -229,8 +230,8 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>10.5047</td>
-<td>51.5647</td>
+<td>5.93 / 5.93</td>
+<td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
 </tr>
@@ -248,19 +249,19 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>[Latest] Further upgraded based on PP-OCRv3, with improved accuracy under comparable speed conditions.</td>
+<td> 70.39</td>
+<td>4.81 / 4.81</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Ultra-lightweight model, supporting English and numeric recognition.</td>
+<td>70.69</td>
+<td>5.44 / 5.44</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
 </tr>
 </table>
 
@@ -276,85 +277,89 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Korean Recognition</td>
+<td>60.21</td>
+<td>5.40 / 5.40</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Japanese Recognition</td>
+<td>45.69</td>
+<td>5.70 / 5.70</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Traditional Chinese Recognition</td>
+<td>82.06</td>
+<td>5.90 / 5.90</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Telugu Recognition</td>
+<td>95.88</td>
+<td>5.42 / 5.42</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Kannada Recognition</td>
+<td>96.96</td>
+<td>5.25 / 5.25</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Tamil Recognition</td>
+<td>76.83</td>
+<td>5.23 / 5.23</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Latin Recognition</td>
+<td>76.93</td>
+<td>5.20 / 5.20</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Arabic Script Recognition</td>
+<td>73.55</td>
+<td>5.35 / 5.35</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Cyrillic Script Recognition</td>
+<td>94.28</td>
+<td>5.23 / 5.23</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Devanagari Script Recognition</td>
+<td>96.44</td>
+<td>5.22 / 5.22</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
+
+</details>
+
 <p><b>Text Line Orientation Classification Module (Optional):</b></p>
 <table>
 <thead>
@@ -408,7 +413,6 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 | Regular Mode| FP32 Precision / No TRT Acceleration   | FP32 Precision / 8 Threads | PaddleInference                                 |
 | High-Performance Mode | Optimal combination of pre-selected precision types and acceleration strategies | FP32 Precision / 8 Threads | Pre-selected optimal backend (Paddle/OpenVINO/TRT, etc.) |
 
-</details>
 
 ## 2. Quick Start
 All model pipelines provided by PaddleX can be quickly experienced. You can experience the effect of the general OCR pipeline on the community platform, or you can use the command line or Python locally to experience the effect of the general OCR pipeline.
@@ -449,6 +453,7 @@ After running, the results will be printed to the terminal as follows:
        [82, 33],
        [ 3, 33]], dtype=int16), ...], 'rec_boxes': array([[  3,  10,  82,  33], ...], dtype=int16)}}
 ```
+The explanation of the running result parameters can refer to the result interpretation in [2.2.2 Python Script Integration](#222-python-script-integration).
 
 The visualized results are saved under `save_path`, and the OCR visualization results are as follows:
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/ocr/03.png"/>
@@ -735,6 +740,8 @@ In the above Python script, the following steps are executed:
 
     - `input_path`: `(str)` The input path of the image to be predicted
 
+    - `page_index`: `(Union[int, None])` If the input is a PDF file, this indicates the current page number of the PDF. Otherwise, it is `None`
+
     - `model_settings`: `(Dict[str, bool])` The model parameters required for the pipeline configuration
 
         - `use_doc_preprocessor`: `(bool)` Controls whether to enable the document preprocessing sub-line
@@ -771,7 +778,7 @@ In the above Python script, the following steps are executed:
 
     - `rec_boxes`: `(numpy.ndarray)` An array of rectangular bounding boxes for detection boxes, with a shape of (n, 4) and dtype int16. Each row represents the [x_min, y_min, x_max, y_max] coordinates of a rectangle, where (x_min, y_min) is the top-left corner and (x_max, y_max) is the bottom-right corner
 
-- Calling the `save_to_json()` method will save the above content to the specified `save_path`. If a directory is specified, the saved path will be `save_path/{your_img_basename}.json`. If a file is specified, it will be saved directly to that file. Since JSON files do not support saving numpy arrays, the `numpy.array` type will be converted to a list format.
+- Calling the `save_to_json()` method will save the above content to the specified `save_path`. If a directory is specified, the saved path will be `save_path/{your_img_basename}_res.json`. If a file is specified, it will be saved directly to that file. Since JSON files do not support saving numpy arrays, the `numpy.array` type will be converted to a list format.
 
 - Calling the `save_to_img()` method will save the visualization results to the specified `save_path`. If a directory is specified, the saved path will be `save_path/{your_img_basename}_ocr_res_img.{your_img_extension}`. If a file is specified, it will be saved directly to that file. (Since the pipeline usually contains multiple result images, it is not recommended to specify a specific file path directly, as multiple images will be overwritten and only the last image will be retained)
 
@@ -1139,24 +1146,24 @@ SubPipelines:
       DocOrientationClassify:
         module_name: doc_text_orientation
         model_name: PP-LCNet_x1_0_doc_ori
-        model_dir: null # 替换为微调后的文档图像方向分类模型权重路径
+        model_dir: null # Replace with the path to the fine-tuned document image orientation classification model weights.
     ...
 
 SubModules:
   TextDetection:
     module_name: text_detection
     model_name: PP-OCRv4_mobile_det
-    model_dir: null # 替换为微调后的文本检测模型权重路径
+    model_dir: null # Replace with the path to the fine-tuned text detection model weights.
     ...
   TextLineOrientation:
     module_name: textline_orientation
     model_name: PP-LCNet_x0_25_textline_ori
-    model_dir: null  # 替换为微调后的文本行方向分类模型权重路径
+    model_dir: null  # Replace with the path to the fine-tuned textline orientation classification model weights.
     batch_size: 1
   TextRecognition:
     module_name: text_recognition
     model_name: PP-OCRv4_mobile_rec
-    model_dir: null  # 替换为微调后的文本识别模型权重路径
+    model_dir: null  # Replace with the path to the fine-tuned text recognition model weights.
     batch_size: 1
 ```
 
@@ -1176,6 +1183,7 @@ paddlex --pipeline OCR \
         --save_path ./output \
         --device npu:0
 ```
+Of course, you can also specify the hardware device when calling `create_pipeline()` or `predict()` in a Python script.
 
 If you want to use the General OCR pipeline on more types of hardware, please refer to the [PaddleX Multi-Hardware Usage Guide](../../../other_devices_support/multi_devices_use_guide.en.md).
 </details>

docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md

@@ -58,7 +58,7 @@ OCR（光学字符识别，Optical Character Recognition）是一种将图像中
 </tr>
 </tbody>
 </table>
-<b>注：模型的精度指标测量自 <a href="https://www3.cs.stonybrook.edu/~cvl/docunet.html">DocUNet benchmark</a>。</b>
+
 <p><b>文本检测模块：</b></p>
 <table>
 <thead>
@@ -373,7 +373,7 @@ devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>
 </tr>
 </table>
-
+</details>
 <p><b>文本行方向分类模块（可选）：</b></p>
 <table>
 <thead>
@@ -424,7 +424,7 @@ devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 | 常规模式    | FP32精度 / 无TRT加速             | FP32精度 / 8线程       | PaddleInference                             |
 | 高性能模式  | 选择先验精度类型和加速策略的最优组合         | FP32精度 / 8线程       | 选择先验最优后端（Paddle/OpenVINO/TRT等） |
 
-</details>
+
 
 ## 2. 快速开始
 PaddleX 所提供的模型产线均可以快速体验效果，你可以在星河社区线体验通用 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

@@ -11,6 +11,7 @@ Formula recognition is a technology that automatically identifies and extracts L
 The formula recognition pipeline is designed to solve formula recognition tasks by extracting formula information from images and outputting it in LaTeX source code format. This pipeline integrates the advanced formula recognition model PP-FormulaNet developed by the PaddlePaddle Vision Team and the well-known formula recognition model UniMERNet. It is an end-to-end formula recognition system that supports the recognition of simple printed formulas, complex printed formulas, and handwritten formulas. Additionally, it includes functions for image orientation correction and distortion correction. Based on this pipeline, precise formula content prediction can be achieved, covering various application scenarios in education, research, finance, manufacturing, and other fields. The pipeline also provides flexible deployment options, supporting multiple hardware devices and programming languages. Moreover, it offers the capability for secondary development. You can train and optimize the pipeline on your own dataset, and the trained model can be seamlessly integrated.
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/formula_recognition/03.png" style="width: 70%"/>
+
 <b>The formula recognition pipeline includes a mandatory formula recognition module,</b> as well as optional layout detection, document image orientation classification, and text image unwarping modules. The document image orientation classification module and the text image unwarping module are integrated into the formula recognition pipeline as a document preprocessing sub-pipeline. Each module contains multiple models, and you can choose the model based on the benchmark test data below.
 
 <b>If you prioritize model accuracy, choose a model with higher precision; if you care more about inference speed, choose a faster model; if you are concerned about model storage size, choose a smaller model.</b>
@@ -38,6 +39,8 @@ The formula recognition pipeline is designed to solve formula recognition tasks
 </tbody>
 </table>
 
+<p><b>Text Image Correction Module (Optional):</b></p>
+
 <table>
 <thead>
 <tr>
@@ -223,36 +226,42 @@ The formula recognition pipeline is designed to solve formula recognition tasks
 </tr>
 </tbody>
 </table>
+</details>
 
 <p><b>Formula Recognition Module </b></p>
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Avg-BLEU(%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <td>UniMERNet</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">Training Model</a></td>
 <td>86.13</td>
-<td>2266.96</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td>UniMERNet is a formula recognition model developed by Shanghai AI Lab. It uses Donut Swin as the encoder and MBartDecoder as the decoder. The model is trained on a dataset of one million samples, including simple formulas, complex formulas, scanned formulas, and handwritten formulas, significantly improving the recognition accuracy of real-world formulas.</td>
 <tr>
 <td>PP-FormulaNet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">Training Model</a></td>
 <td>87.12</td>
-<td>202.25</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td rowspan="2">PP-FormulaNet is an advanced formula recognition model developed by the Baidu PaddlePaddle Vision Team. The PP-FormulaNet-S version uses PP-HGNetV2-B4 as its backbone network. Through parallel masking and model distillation techniques, it significantly improves inference speed while maintaining high recognition accuracy, making it suitable for applications requiring fast inference. The PP-FormulaNet-L version, on the other hand, uses Vary_VIT_B as its backbone network and is trained on a large-scale formula dataset, showing significant improvements in recognizing complex formulas compared to PP-FormulaNet-S.</td>
 </tr>
 <td>PP-FormulaNet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">Training Model</a></td>
 <td>92.13</td>
-<td>1976.52</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <tr>
 <td>LaTeX_OCR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">Training Model</a></td>
 <td>71.63</td>
-<td>-</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td>LaTeX-OCR is a formula recognition algorithm based on an autoregressive large model. It uses Hybrid ViT as the backbone network and a transformer as the decoder, significantly improving the accuracy of formula recognition.</td>
 </tr>
@@ -282,7 +291,7 @@ The formula recognition pipeline is designed to solve formula recognition tasks
 | Regular Mode| FP32 Precision / No TRT Acceleration   | FP32 Precision / 8 Threads | PaddleInference                                 |
 | High-Performance Mode | Optimal combination of pre-selected precision types and acceleration strategies | FP32 Precision / 8 Threads | Pre-selected optimal backend (Paddle/OpenVINO/TRT, etc.) |
 
-</details>
+
 
 ## 2. Quick Start
 All model pipelines provided by PaddleX can be quickly experienced. You can experience the effect of the formula recognition pipeline on the community platform, or you can use the command line or Python locally to experience the effect of the formula recognition pipeline.
@@ -327,6 +336,7 @@ The explanation of the running result parameters can refer to the result interpr
 The visualization results are saved under `save_path`, where the visualization result of formula recognition is as follows:
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/formula_recognition/04.png" style="width: 70%"/>
+
 <b>If you need to visualize the formula recognition pipeline, you need to run the following command to install the LaTeX rendering environment. Currently, visualization of the formula recognition pipeline only supports the Ubuntu environment, and other environments are not supported. For complex formulas, the LaTeX result may contain some advanced representations that may not be successfully displayed in environments such as Markdown:</b>
 
 ```bash

docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md

@@ -12,6 +12,7 @@ comments: true
 
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/formula_recognition/03.png" style="width: 70%"/>
+
 <b>公式识别产线中包含必选的公式识别模块，</b>以及可选的版面区域检测模块、文档图像方向分类模块和文本图像矫正模块。其中，文档图像方向分类模块和文本图像矫正模块作为文档预处理子产线被集成到公式识别产线中。每个模块都包含多个模型，您可以根据下方的基准测试数据选择使用的模型。
 
 <b>如果您更注重模型的精度，请选择精度较高的模型；如果您更在意模型的推理速度，请选择推理速度较快的模型；如果您关注模型的存储大小，请选择存储体积较小的模型。</b>
@@ -226,36 +227,43 @@ comments: true
 </tr>
 </tbody>
 </table>
+</details>
 
 <p><b>公式识别模块：</b></p>
 <table>
 <tr>
 <th>模型</th><th>模型下载链接</th>
 <th>Avg-BLEU(%)</th>
-<th>GPU推理耗时 (ms)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
 <th>模型存储大小 (M)</th>
 <th>介绍</th>
 </tr>
 <td>UniMERNet</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">训练模型</a></td>
 <td>86.13</td>
-<td>2266.96</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td>UniMERNet是由上海AI Lab研发的一款公式识别模型。该模型采用Donut Swin作为编码器，MBartDecoder作为解码器，并通过在包含简单公式、复杂公式、扫描捕捉公式和手写公式在内的一百万数据集上进行训练，大幅提升了模型对真实场景公式的识别准确率</td>
 <tr>
 <td>PP-FormulaNet-S</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">训练模型</a></td>
 <td>87.12</td>
-<td>202.25</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td rowspan="2">PP-FormulaNet 是由百度飞桨视觉团队开发的一款先进的公式识别模型，支持5万个常见LateX源码词汇的识别。PP-FormulaNet-S 版本采用了 PP-HGNetV2-B4 作为其骨干网络，通过并行掩码和模型蒸馏等技术，大幅提升了模型的推理速度，同时保持了较高的识别精度，适用于简单印刷公式、跨行简单印刷公式等场景。而 PP-FormulaNet-L 版本则基于 Vary_VIT_B 作为骨干网络，并在大规模公式数据集上进行了深入训练，在复杂公式的识别方面，相较于PP-FormulaNet-S表现出显著的提升，适用于简单印刷公式、复杂印刷公式、手写公式等场景。 </td>
+
 </tr>
 <td>PP-FormulaNet-L</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">训练模型</a></td>
 <td>92.13</td>
-<td>1976.52</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <tr>
 <td>LaTeX_OCR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">训练模型</a></td>
 <td>71.63</td>
-<td>-</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td>LaTeX-OCR是一种基于自回归大模型的公式识别算法，通过采用 Hybrid ViT 作为骨干网络，transformer作为解码器，显著提升了公式识别的准确性。</td>
 </tr>
@@ -285,7 +293,7 @@ comments: true
 | 常规模式    | FP32精度 / 无TRT加速             | FP32精度 / 8线程       | PaddleInference                             |
 | 高性能模式  | 选择先验精度类型和加速策略的最优组合         | FP32精度 / 8线程       | 选择先验最优后端（Paddle/OpenVINO/TRT等） |
 
-</details>
+
 
 ## 2. 快速开始
 PaddleX 所提供的模型产线均可以快速体验效果，你可以在星河社区线体验公式识别产线的效果，也可以在本地使用命令行或 Python 体验公式识别产线的效果。
@@ -329,7 +337,9 @@ paddlex --pipeline formula_recognition \
 可视化结果保存在`save_path`下，其中公式识别的可视化结果如下：
 
 <img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/pipelines/formula_recognition/04.png" style="width: 70%"/>
+
 <b> 如果您需要对公式识别产线进行可视化，需要运行如下命令来对LaTeX渲染环境进行安装。目前公式识别产线可视化只支持Ubuntu环境，其他环境暂不支持。对于复杂公式，LaTeX 结果可能包含部分高级的表示，Markdown等环境中未必可以成功显示：</b>
+
 ```bash
 sudo apt-get update
 sudo apt-get install texlive texlive-latex-base texlive-latex-extra -y

docs/pipeline_usage/tutorials/video_pipelines/video_classification.md

@@ -206,7 +206,8 @@ for res in output:
 
 </tbody>
 </table>
-（3）对预测结果进行处理，每个样本的预测结果均为对应的Result对象，且支持打印、保存为图片、保存为`json`文件的操作:
+
+（3）对预测结果进行处理，每个样本的预测结果均为对应的Result对象，且支持打印、保存为图片、保存为 `json` 文件的操作:
 
 <table>
 <thead>

docs/support_list/models_list.en.md

@@ -2178,6 +2178,9 @@ PaddleX includes multiple production lines, each containing several modules, and
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 </tr>
 </table>
+
+<p><b>Note: The evaluation set for the above accuracy metrics is an English dataset built by PaddleX. All model GPU inference times are based on NVIDIA Tesla T4 machines, with precision type FP32. CPU inference speed is based on Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz, with 8 threads, and precision type FP32.</b></p>
+
 <b>Multilingual Recognition Model</b>
 <table>
 <tr>
@@ -2287,36 +2290,41 @@ PaddleX includes multiple production lines, each containing several modules, and
 <table>
 <tr>
 <th>Model</th>
-<th>Avg-BLEU</th>
-<th>GPU Inference Time (ms)</th>
+<th>Avg-BLEU(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>yaml File</th>
 <th>Model Download Link</th>
 </tr>
 <td>UniMERNet</td>
-<td>0.8613</td>
-<td>2266.96</td>
+<td>86.13</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/UniMERNet.yaml">UniMERNet.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">Training Model</a></td>
 <tr>
 <td>PP-FormulaNet-S</td>
-<td>0.8712</td>
-<td>202.25</td>
+<td>87.12</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/PP-FormulaNet-S.yaml">PP-FormulaNet-S.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">Training Model</a></td>
 </tr>
 <td>PP-FormulaNet-L</td>
-<td>0.9213</td>
-<td>1976.52</td>
+<td>92.13</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/PP-FormulaNet-L.yaml">PP-FormulaNet-L.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">Training Model</a></td>
 <tr>
 <td>LaTeX_OCR_rec</td>
-<td>0.7163</td>
-<td>-</td>
+<td>71.63</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/LaTeX_OCR_rec.yaml">LaTeX_OCR_rec.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">Training Model</a></td>

docs/support_list/models_list.md

@@ -2042,6 +2042,8 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 </table>
 
+<p><b>注：以上精度指标的评估集是 PaddleX 自建的英文数据集。 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>
+
 * <b>多语言识别模型</b>
 <table>
 <tr>
@@ -2161,36 +2163,41 @@ devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 <table>
 <tr>
 <th>模型</th>
-<th>Avg-BLEU</th>
-<th>GPU推理耗时 (ms)</th>
+<th>Avg-BLEU(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
 <th>模型存储大小 (M)</th>
 <th>yaml 文件</th>
 <th>模型下载链接</th>
 </tr>
 <td>UniMERNet</td>
-<td>0.8613</td>
-<td>2266.96</td>
+<td>86.13</td>
+<td>2266.96/-</td>
+<td>-/-</td>
 <td>1.4 G</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/UniMERNet.yaml">UniMERNet.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/UniMERNet_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/UniMERNet_pretrained.pdparams">训练模型</a></td>
 <tr>
 <td>PP-FormulaNet-S</td>
-<td>0.8712</td>
-<td>202.25</td>
+<td>87.12</td>
+<td>202.25/-</td>
+<td>-/-</td>
 <td>167.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/PP-FormulaNet-S.yaml">PP-FormulaNet-S.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-S_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-S_pretrained.pdparams">训练模型</a></td>
 </tr>
 <td>PP-FormulaNet-L</td>
-<td>0.9213</td>
-<td>1976.52</td>
+<td>92.13</td>
+<td>1976.52/-</td>
+<td>-/-</td>
 <td>535.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/PP-FormulaNet-L.yaml">PP-FormulaNet-L.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-FormulaNet-L_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-FormulaNet-L_pretrained.pdparams">训练模型</a></td>
 <tr>
 <td>LaTeX_OCR_rec</td>
-<td>0.7163</td>
-<td>-</td>
+<td>71.63</td>
+<td>-/-</td>
+<td>-/-</td>
 <td>89.7 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/formula_recognition/LaTeX_OCR_rec.yaml">LaTeX_OCR_rec.yaml</a></td>
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/LaTeX_OCR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/LaTeX_OCR_rec_pretrained.pdparams">训练模型</a></td>
@@ -2769,4 +2776,4 @@ devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 <td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/YOWO_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/YOWO_pretrained.pdparams">训练模型</a></td>
 </tr>
 </table>
-<p><b>注：以上精度指标为 <a href="http://www.thumos.info/download.html">UCF101-24</a> test数据集上的测试指标Frame-mAP (@ IoU 0.5)。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>注：以上精度指标为 <a href="http://www.thumos.info/download.html">UCF101-24</a> test数据集上的测试指标Frame-mAP (@ IoU 0.5)。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>