Browse Source

update doc about CINN

gaotingquan 7 months ago
parent
commit
c6094e3b3e
68 changed files with 146 additions and 92 deletions
  1. 2 1
      docs/module_usage/tutorials/cv_modules/3d_bev_detection.en.md
  2. 2 1
      docs/module_usage/tutorials/cv_modules/3d_bev_detection.md
  3. 2 2
      docs/module_usage/tutorials/cv_modules/anomaly_detection.en.md
  4. 2 1
      docs/module_usage/tutorials/cv_modules/anomaly_detection.md
  5. 2 2
      docs/module_usage/tutorials/cv_modules/face_detection.en.md
  6. 2 1
      docs/module_usage/tutorials/cv_modules/face_detection.md
  7. 2 1
      docs/module_usage/tutorials/cv_modules/face_feature.en.md
  8. 2 1
      docs/module_usage/tutorials/cv_modules/face_feature.md
  9. 2 1
      docs/module_usage/tutorials/cv_modules/human_detection.en.md
  10. 2 1
      docs/module_usage/tutorials/cv_modules/human_detection.md
  11. 2 1
      docs/module_usage/tutorials/cv_modules/human_keypoint_detection.en.md
  12. 2 1
      docs/module_usage/tutorials/cv_modules/human_keypoint_detection.md
  13. 3 1
      docs/module_usage/tutorials/cv_modules/image_classification.en.md
  14. 2 1
      docs/module_usage/tutorials/cv_modules/image_classification.md
  15. 2 1
      docs/module_usage/tutorials/cv_modules/image_feature.en.md
  16. 2 1
      docs/module_usage/tutorials/cv_modules/image_feature.md
  17. 2 1
      docs/module_usage/tutorials/cv_modules/image_multilabel_classification.en.md
  18. 2 1
      docs/module_usage/tutorials/cv_modules/image_multilabel_classification.md
  19. 3 2
      docs/module_usage/tutorials/cv_modules/instance_segmentation.en.md
  20. 2 1
      docs/module_usage/tutorials/cv_modules/instance_segmentation.md
  21. 2 1
      docs/module_usage/tutorials/cv_modules/mainbody_detection.en.md
  22. 2 1
      docs/module_usage/tutorials/cv_modules/mainbody_detection.md
  23. 3 2
      docs/module_usage/tutorials/cv_modules/object_detection.en.md
  24. 2 1
      docs/module_usage/tutorials/cv_modules/object_detection.md
  25. 2 1
      docs/module_usage/tutorials/cv_modules/pedestrian_attribute_recognition.en.md
  26. 2 1
      docs/module_usage/tutorials/cv_modules/pedestrian_attribute_recognition.md
  27. 2 1
      docs/module_usage/tutorials/cv_modules/rotated_object_detection.en.md
  28. 2 1
      docs/module_usage/tutorials/cv_modules/rotated_object_detection.md
  29. 2 2
      docs/module_usage/tutorials/cv_modules/semantic_segmentation.en.md
  30. 2 1
      docs/module_usage/tutorials/cv_modules/semantic_segmentation.md
  31. 2 1
      docs/module_usage/tutorials/cv_modules/small_object_detection.en.md
  32. 2 1
      docs/module_usage/tutorials/cv_modules/small_object_detection.md
  33. 2 1
      docs/module_usage/tutorials/cv_modules/vehicle_attribute_recognition.en.md
  34. 2 1
      docs/module_usage/tutorials/cv_modules/vehicle_attribute_recognition.md
  35. 2 1
      docs/module_usage/tutorials/cv_modules/vehicle_detection.en.md
  36. 2 1
      docs/module_usage/tutorials/cv_modules/vehicle_detection.md
  37. 2 4
      docs/module_usage/tutorials/ocr_modules/doc_img_orientation_classification.en.md
  38. 2 1
      docs/module_usage/tutorials/ocr_modules/doc_img_orientation_classification.md
  39. 2 1
      docs/module_usage/tutorials/ocr_modules/formula_recognition.en.md
  40. 2 1
      docs/module_usage/tutorials/ocr_modules/formula_recognition.md
  41. 2 2
      docs/module_usage/tutorials/ocr_modules/layout_detection.en.md
  42. 2 1
      docs/module_usage/tutorials/ocr_modules/layout_detection.md
  43. 2 2
      docs/module_usage/tutorials/ocr_modules/seal_text_detection.en.md
  44. 2 1
      docs/module_usage/tutorials/ocr_modules/seal_text_detection.md
  45. 2 1
      docs/module_usage/tutorials/ocr_modules/table_cells_detection.en.md
  46. 2 1
      docs/module_usage/tutorials/ocr_modules/table_cells_detection.md
  47. 2 1
      docs/module_usage/tutorials/ocr_modules/table_classification.en.md
  48. 2 1
      docs/module_usage/tutorials/ocr_modules/table_classification.md
  49. 3 2
      docs/module_usage/tutorials/ocr_modules/table_structure_recognition.en.md
  50. 2 1
      docs/module_usage/tutorials/ocr_modules/table_structure_recognition.md
  51. 2 1
      docs/module_usage/tutorials/ocr_modules/text_detection.en.md
  52. 2 1
      docs/module_usage/tutorials/ocr_modules/text_detection.md
  53. 1 1
      docs/module_usage/tutorials/ocr_modules/text_image_unwarping.md
  54. 2 2
      docs/module_usage/tutorials/ocr_modules/text_recognition.en.md
  55. 2 2
      docs/module_usage/tutorials/ocr_modules/text_recognition.md
  56. 2 1
      docs/module_usage/tutorials/ocr_modules/textline_orientation_classification.en.md
  57. 3 2
      docs/module_usage/tutorials/ocr_modules/textline_orientation_classification.md
  58. 2 1
      docs/module_usage/tutorials/speech_modules/multilingual_speech_recognition.en.md
  59. 2 2
      docs/module_usage/tutorials/time_series_modules/time_series_anomaly_detection.en.md
  60. 3 2
      docs/module_usage/tutorials/time_series_modules/time_series_anomaly_detection.md
  61. 2 2
      docs/module_usage/tutorials/time_series_modules/time_series_classification.en.md
  62. 3 2
      docs/module_usage/tutorials/time_series_modules/time_series_classification.md
  63. 2 2
      docs/module_usage/tutorials/time_series_modules/time_series_forecasting.en.md
  64. 3 2
      docs/module_usage/tutorials/time_series_modules/time_series_forecasting.md
  65. 3 2
      docs/module_usage/tutorials/video_modules/video_classification.en.md
  66. 2 1
      docs/module_usage/tutorials/video_modules/video_classification.md
  67. 3 3
      docs/module_usage/tutorials/video_modules/video_detection.en.md
  68. 3 2
      docs/module_usage/tutorials/video_modules/video_detection.md

+ 2 - 1
docs/module_usage/tutorials/cv_modules/3d_bev_detection.en.md

@@ -393,7 +393,8 @@ python main.py -c paddlex/configs/modules/3d_bev_detection/BEVFusion.yaml \
 * Specify the path of the model's `.yaml` configuration file (here it is `bevf_pp_2x8_1x_nusc.yaml`. When training other models, you need to specify the corresponding configuration file. The correspondence between models and configuration files can be found in [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding model task module [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding model task module [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Information (Click to Expand)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/cv_modules/3d_bev_detection.md

@@ -406,7 +406,8 @@ python main.py -c paddlex/configs/modules/3d_bev_detection/BEVFusion.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`bevf_pp_2x8_1x_nusc.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/cv_modules/anomaly_detection.en.md

@@ -344,8 +344,8 @@ The steps required are:
 * Specify the path to the `.yaml` configuration file of the model (here it is `STFPM.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/cv_modules/anomaly_detection.md

@@ -347,7 +347,8 @@ python main.py -c paddlex/configs/modules/image_anomaly_detection/STFPM.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`STFPM.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/cv_modules/face_detection.en.md

@@ -453,8 +453,8 @@ The steps required are:
 * Specify the path to the `.yaml` configuration file of the model (here it is `PicoDet_LCNet_x2_5_face.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/face_detection.md

@@ -443,7 +443,8 @@ python main.py -c paddlex/configs/modules/face_detection/PicoDet_LCNet_x2_5_face
 * 指定模型的`.yaml` 配置文件路径(此处为`PicoDet_LCNet_x2_5_face.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/face_feature.en.md

@@ -402,7 +402,8 @@ The steps required are:
 * Specify the path to the `.yaml` configuration file for the model (here it is `MobileFaceNet.yaml`)
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or by appending parameters in the command line. For example, to specify the first two GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or by appending parameters in the command line. For example, to specify the first two GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/face_feature.md

@@ -425,7 +425,8 @@ python main.py -c paddlex/configs/modules/face_feature/MobileFaceNet.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`MobileFaceNet.yaml`)
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/human_detection.en.md

@@ -424,7 +424,8 @@ The steps required are:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-YOLOE-S_human.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/human_detection.md

@@ -422,7 +422,8 @@ python main.py -c paddlex/configs/modules/human_detection/PP-YOLOE-S_human.yaml
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-YOLOE-S_human.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/human_keypoint_detection.en.md

@@ -438,7 +438,8 @@ Similar to model training, the process involves the following steps:
 * Specify the path to the `.yaml` configuration file for the model(here it's `MobileFaceNet.yaml`)
 * Set the mode to model evaluation: `-o Global.mode=evaluate`
 * Specify the path to the validation dataset: `-o Global.dataset_dir`
-Other related parameters can be configured by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration file. For detailed information, please refer to [PaddleX Common Configuration Parameters for Models](../../instructions/config_parameters_common.en.md)。
+* Other related parameters can be configured by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration file. For detailed information, please refer to [PaddleX Common Configuration Parameters for Models](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details>
 <summary>👉 <b>More Details (Click to Expand)</b></summary>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/human_keypoint_detection.md

@@ -454,7 +454,8 @@ python main.py -c paddlex/configs/modules/keypoint_detection/PP-TinyPose_128x96.
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-TinyPose_128x96.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details>
   <summary>👉 <b>更多说明(点击展开)</b></summary>

+ 3 - 1
docs/module_usage/tutorials/cv_modules/image_classification.en.md

@@ -1062,7 +1062,9 @@ Similar to model training, the following steps are required:
 
 * Specify the path of the model's `.yaml` configuration file (here it is `PP-LCNet_x1_0.yaml`)
 * Specify the mode as model evaluation: `-o Global.mode=evaluate`
-* Specify the path of the validation dataset: `-o Global.dataset_dir`. Other related parameters can be set by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration. Other related parameters can be set by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration file. For details, please refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* Specify the path of the validation dataset: `-o Global.dataset_dir`
+* Other related parameters can be set by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration. Other related parameters can be set by modifying the fields under `Global` and `Evaluate` in the `.yaml` configuration file. For details, please refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <p>When evaluating the model, you need to specify the model weight 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_model/best_model.pdparams</code>.</p>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/image_classification.md

@@ -1051,7 +1051,8 @@ python main.py -c paddlex/configs/modules/image_classification/PP-LCNet_x1_0.yam
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/image_feature.en.md

@@ -440,7 +440,8 @@ The following steps are required:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-ShiTuV2_rec.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`.
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/image_feature.md

@@ -440,7 +440,8 @@ python main.py -c paddlex/configs/modules/image_feature/PP-ShiTuV2_rec.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-ShiTuV2_rec.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/image_multilabel_classification.en.md

@@ -503,7 +503,8 @@ Similar to model training, the following steps are required:
 * Specify the `.yaml` configuration file path for the model (here it's `PP-LCNet_x1_0_ML.yaml`)
 * Specify the mode as model evaluation: `-o Global.mode=evaluate`
 * Specify the path to the validation dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Evaluate` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Evaluate` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <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. If you need to change it, simply append the command line parameter to set it, such as <code>-o Evaluate.weight_path=./output/best_model/best_model.pdparams</code>.</p>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/image_multilabel_classification.md

@@ -493,7 +493,8 @@ python main.py -c paddlex/configs/modules/image_multilabel_classification/PP-LCN
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0_ML.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 3 - 2
docs/module_usage/tutorials/cv_modules/instance_segmentation.en.md

@@ -572,8 +572,9 @@ The following steps are required:
 
 * Specify the path to the `.yaml` configuration file of the model (here it is `Mask-RT-DETR-L.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
-* Specify the path to the training dataset: `-o Global.dataset_dir`.
-Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify the first 2 GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration File Parameters Instructions](../../instructions/config_parameters_common.en.md).
+* Specify the path to the training dataset: `-o Global.dataset_dir`
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify the first 2 GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration File Parameters Instructions](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/instance_segmentation.md

@@ -577,7 +577,8 @@ python main.py -c paddlex/configs/modules/instance_segmentation/Mask-RT-DETR-L.y
 * 指定模型的`.yaml` 配置文件路径(此处为 `Mask-RT-DETR-L.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/mainbody_detection.en.md

@@ -415,7 +415,8 @@ The steps required are:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-ShiTuV2_det.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/mainbody_detection.md

@@ -410,7 +410,8 @@ python main.py -c paddlex/configs/modules/mainbody_detection/PP-ShiTuV2_det.yaml
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-ShiTuV2_det.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 3 - 2
docs/module_usage/tutorials/cv_modules/object_detection.en.md

@@ -790,8 +790,9 @@ The following steps are required:
 
 * Specify the `.yaml` configuration file path for the model (here it is `PicoDet-S.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Set the mode to model training: `-o Global.mode=train`
-* Specify the path to the training dataset: `-o Global.dataset_dir`.
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* Specify the path to the training dataset: `-o Global.dataset_dir`
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/object_detection.md

@@ -809,7 +809,8 @@ python main.py -c paddlex/configs/modules/object_detection/PicoDet-S.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`PicoDet-S.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/pedestrian_attribute_recognition.en.md

@@ -480,7 +480,8 @@ Similar to model training, the following steps are required:
 * Specify the path to the model's `.yaml` configuration file (here it is `PP-LCNet_x1_0_pedestrian_attribute.yaml`)
 * Specify the mode as model evaluation: `-o Global.mode=evaluate`
 * Specify the path to the validation dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Evaluate` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Evaluate` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <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_model/best_model.pdparams</code>.</p>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/pedestrian_attribute_recognition.md

@@ -448,7 +448,8 @@ python main.py -c paddlex/configs/modules/pedestrian_attribute_recognition/PP-LC
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0_pedestrian_attribute.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/rotated_object_detection.en.md

@@ -451,7 +451,8 @@ The following steps are required:
 * Specify the path of the model's `.yaml` configuration file (here it is `PP-YOLOE-R-L.yaml`. When training other models, you need to specify the corresponding configuration file. The correspondence between models and configuration files can be found in [PaddleX Model List (CPU/GPU))](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the fields under Global and Train in the `.yaml` configuration file, or by adding parameters in the command line. For example, specify the first 2 GPU cards for training: `-o Global.device=gpu:0,1`; set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and detailed explanations, please refer to the configuration file instructions for the corresponding task module [PaddleX Common Model Configuration File Parameter Instructions.](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the fields under Global and Train in the `.yaml` configuration file, or by adding parameters in the command line. For example, specify the first 2 GPU cards for training: `-o Global.device=gpu:0,1`; set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and detailed explanations, please refer to the configuration file instructions for the corresponding task module [PaddleX Common Model Configuration File Parameter Instructions.](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Explanations (Click to Expand)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/cv_modules/rotated_object_detection.md

@@ -446,7 +446,8 @@ python main.py -c paddlex/configs/modules/rotated_object_detection/PP-YOLOE-R-L.
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-YOLOE-R-L.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/cv_modules/semantic_segmentation.en.md

@@ -624,8 +624,8 @@ You need to follow these steps:
 * Specify the `.yaml` configuration file path for the model (here it's `PP-LiteSeg-T.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/semantic_segmentation.md

@@ -616,7 +616,8 @@ python main.py -c paddlex/configs/modules/semantic_segmentation/PP-LiteSeg-T.yam
 * 指定模型的.yaml 配置文件路径(此处为 `PP-LiteSeg-T.yam`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/small_object_detection.en.md

@@ -463,7 +463,8 @@ The steps required are:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-YOLOE_plus_SOD-S.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/small_object_detection.md

@@ -457,7 +457,8 @@ python main.py -c paddlex/configs/modules/small_object_detection/PP-YOLOE_plus_S
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-YOLOE_plus_SOD-S.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/vehicle_attribute_recognition.en.md

@@ -431,7 +431,8 @@ The steps required are:
 * Specify the path to the model's `.yaml` configuration file (here it's `PP-LCNet_x1_0_vehicle_attribute.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/vehicle_attribute_recognition.md

@@ -428,7 +428,8 @@ python main.py -c paddlex/configs/modules/vehicle_attribute_recognition/PP-LCNet
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0_vehicle_attribute.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/vehicle_detection.en.md

@@ -419,7 +419,8 @@ The steps required are:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-YOLOE-S_vehicle.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/cv_modules/vehicle_detection.md

@@ -416,7 +416,8 @@ python main.py -c paddlex/configs/modules/vehicle_detection/PP-YOLOE-S_vehicle.y
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-YOLOE-S_vehicle.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 4
docs/module_usage/tutorials/ocr_modules/doc_img_orientation_classification.en.md

@@ -475,12 +475,10 @@ python main.py -c paddlex/configs/modules/doc_text_orientation/PP-LCNet_x1_0_doc
 Similar to model training and evaluation, the following steps are required:
 
 * Specify the `.yaml` configuration file path of the model (here it's `PP-LCNet_x1_0_doc_ori.yaml`)
-
 * Set the mode to model inference prediction: `-o Global.mode=predict`
-
 * Specify the model weights path: -o Predict.model_dir="./output/best_accuracy/inference"
-
-Specify the input data path: `-o Predict.input="..."` Other related parameters can be set by modifying the fields under Global and Predict in the `.yaml` configuration file. For details, refer to PaddleX Common Model Configuration File Parameter Description.
+* Specify the input data path: `-o Predict.input="..."` Other related parameters can be set by modifying the fields under Global and Predict in the `.yaml` configuration file. For details, refer to PaddleX Common Model Configuration File Parameter Description.
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 Alternatively, you can use the PaddleX wheel package for inference, easily integrating the model into your own projects.
 

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/doc_img_orientation_classification.md

@@ -405,7 +405,8 @@ python main.py -c paddlex/configs/modules/doc_text_orientation/PP-LCNet_x1_0_doc
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0_doc_ori.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/formula_recognition.en.md

@@ -455,9 +455,10 @@ The following steps are required:
 * Specify the `.yaml` configuration file path for the model (here it is `PP-FormulaNet-S.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`.
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding task module of the model [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
 *  Except for LaTeX_OCR_rec, the formula recognition models only support exporting models in JSON format. Therefore, during training, you need to set the parameter `FLAGS_json_format_model=1`.
 *  For the PP-FormulaNet-S, PP-FormulaNet-L, and UniMERNet models, additional Linux packages need to be installed during training. The specific command is as follows:
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 ```bash
 sudo apt-get update

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/formula_recognition.md

@@ -451,7 +451,7 @@ FLAGS_json_format_model=1 python main.py -c paddlex/configs/modules/formula_reco
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-FormulaNet-S.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
 * 除 LaTeX_OCR_rec外, 公式识别模型只支持导出json格式的模型,因此训练时需要设置参数`FLAGS_json_format_model=1`。
 * 对于 PP-FormulaNet-S、PP-FormulaNet-L、UniMERNet 模型,在训练还需要安装额外的Linux包,具体命令如下:
 ```bash
@@ -459,6 +459,7 @@ sudo apt-get update
 sudo apt-get install libmagickwand-dev
 python -m pip install Wand
 ```
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/ocr_modules/layout_detection.en.md

@@ -607,8 +607,8 @@ The steps required are:
 * Specify the path to the `.yaml` configuration file of the model (here it is `PP-DocLayout-L.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters for Model Tasks](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/layout_detection.md

@@ -668,7 +668,8 @@ python main.py -c paddlex/configs/modules/layout_detection/PP-DocLayout-L.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-DocLayout-L.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 2
docs/module_usage/tutorials/ocr_modules/seal_text_detection.en.md

@@ -563,8 +563,8 @@ You need to follow these steps:
 * Specify the `.yaml` configuration file path for the model (here it's `PP-OCRv4_server_seal_det.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/seal_text_detection.md

@@ -555,7 +555,8 @@ python main.py -c paddlex/configs/modules/seal_text_detection/PP-OCRv4_server_se
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-OCRv4_server_seal_det.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/table_cells_detection.en.md

@@ -481,7 +481,8 @@ The following steps are required:
 * Specify the path to the model's `.yaml` configuration file (here it is `RT-DETR-L_wired_table_cell_det.yaml`). When training other models, the corresponding configuration file must be specified. The correspondence between models and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by adding parameters in the command line. For example, to train on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding model task module in [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by adding parameters in the command line. For example, to train on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file instructions for the corresponding model task module in [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Information (Click to Expand)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/table_cells_detection.md

@@ -480,7 +480,8 @@ python main.py -c paddlex/configs/modules/table_cells_detection/RT-DETR-L_wired_
 * 指定模型的`.yaml` 配置文件路径(此处为`RT-DETR-L_wired_table_cell_det.yaml`),训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/table_classification.en.md

@@ -387,7 +387,8 @@ python main.py -c paddlex/configs/modules/table_classification/PP-LCNet_x1_0_tab
 * Specify the `.yaml` configuration file path for the model (here it is `PP-LCNet_x1_0_table_cls.yaml`. When training other models, the corresponding configuration file needs to be specified. The correspondence between models and configurations can be found in [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file description of the corresponding model task module [PaddleX General Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file description of the corresponding model task module [PaddleX General Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/table_classification.md

@@ -389,7 +389,8 @@ python main.py -c paddlex/configs/modules/table_classification/PP-LCNet_x1_0_tab
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x1_0_table_cls.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/ocr_modules/table_structure_recognition.en.md

@@ -392,8 +392,9 @@ the following steps are required:
 
 * Specify the path of the model's `.yaml` configuration file (here it is `SLANet.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
-* Specify the path of the training dataset: `-o Global.dataset_dir`. Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
-
+* Specify the path of the training dataset: `-o Global.dataset_dir`.
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/table_structure_recognition.md

@@ -387,7 +387,8 @@ python main.py -c paddlex/configs/modules/table_structure_recognition/SLANet.yam
 * 指定模型的`.yaml` 配置文件路径(此处为`SLANet.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/text_detection.en.md

@@ -508,7 +508,8 @@ The steps required are:
 * Specify the path to the model's `.yaml` configuration file (here it's `PP-OCRv4_mobile_det.yaml`. When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../../module_usage/instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or adjusted by appending parameters in the command line. For example, to specify training on the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration Parameters Documentation](../../../module_usage/instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Information (Click to Expand)</b></summary>
 <ul>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/text_detection.md

@@ -523,7 +523,8 @@ python main.py -c paddlex/configs/modules/text_detection/PP-OCRv4_mobile_det.yam
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-OCRv4_mobile_det.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明 [PaddleX通用模型配置文件参数说明](../../../module_usage/instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明 [PaddleX通用模型配置文件参数说明](../../../module_usage/instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 <ul>

+ 1 - 1
docs/module_usage/tutorials/ocr_modules/text_image_unwarping.md

@@ -16,7 +16,7 @@ comments: true
 <tr>
 <th>模型</th><th>模型下载链接</th>
 <th>CER </th>
-<th>模型存储大小(M)</th>
+<th>模型存储大小(M</th>
 <th>介绍</th>
 </tr>
 </thead>

+ 2 - 2
docs/module_usage/tutorials/ocr_modules/text_recognition.en.md

@@ -644,8 +644,8 @@ The steps required are:
 * Specify the path to the model's `.yaml` configuration file (here it's `PP-OCRv4_mobile_rec.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`.
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
-
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX Common Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Information (Click to Expand)</b></summary>
 <ul>

+ 2 - 2
docs/module_usage/tutorials/ocr_modules/text_recognition.md

@@ -662,8 +662,8 @@ python main.py -c paddlex/configs/modules/text_recognition/PP-OCRv4_mobile_rec.y
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-OCRv4_mobile_rec.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
-
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>

+ 2 - 1
docs/module_usage/tutorials/ocr_modules/textline_orientation_classification.en.md

@@ -404,7 +404,8 @@ The following steps are required:
 * Specify the path to the `.yaml` configuration file for the model (here it is `PP-LCNet_x0_25_textline_ori.yaml`. When training other models, you need to specify the corresponding configuration file. The correspondence between models and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Specify the mode as model training: `-o Global.mode=train`
 * Specify the path to the training dataset: `-o Global.dataset_dir`
-Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file or by appending parameters in the command line. For example, to specify the first two GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file description for the corresponding task module of the model [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file or by appending parameters in the command line. For example, to specify the first two GPUs for training: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file description for the corresponding task module of the model [PaddleX Common Model Configuration Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/ocr_modules/textline_orientation_classification.md

@@ -18,7 +18,7 @@ comments: true
 <th>Top-1 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>
+<th>模型存储大小(M</th>
 <th>介绍</th>
 </tr>
 </thead>
@@ -407,7 +407,8 @@ python main.py -c paddlex/configs/modules/textline_orientation/PP-LCNet_x0_25_te
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-LCNet_x0_25_textline_ori.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 1
docs/module_usage/tutorials/speech_modules/multilingual_speech_recognition.en.md

@@ -223,7 +223,8 @@ the following steps are required for model inference:
 * Specify the `.yaml` configuration file path for the model (here it is `whisper_large.yaml`)
 * Specify the mode as model inference prediction: `-o Global.mode=predict`
 * Specify the input data path: `-o Predict.input="..."`
-Other related parameters can be set by modifying the `Global` and `Predict` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* Other related parameters can be set by modifying the `Global` and `Predict` fields in the `.yaml` configuration file. For details, refer to [PaddleX Common Model Configuration File Parameter Description](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 #### 4.4.2 Model Integration
 Models can be directly integrated into the PaddleX pipelines or into your own projects.

+ 2 - 2
docs/module_usage/tutorials/time_series_modules/time_series_anomaly_detection.en.md

@@ -472,8 +472,8 @@ You need to follow these steps:
 * Specify the `.yaml` configuration file path for the model (here it's `AutoEncoder_ad.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/time_series_modules/time_series_anomaly_detection.md

@@ -17,7 +17,7 @@ comments: true
 <th>precison</th>
 <th>recall</th>
 <th>f1_score</th>
-<th>模型存储大小(M)</th>
+<th>模型存储大小(M</th>
 <th>介绍</th>
 </tr>
 </thead>
@@ -474,7 +474,8 @@ python main.py -c paddlex/configs/modules/ts_anomaly_detection/AutoEncoder_ad.ya
 * 指定模型的`.yaml` 配置文件路径(此处为`AutoEncoder_ad.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/time_series_modules/time_series_classification.en.md

@@ -454,8 +454,8 @@ You need to follow these steps:
 * Specify the `.yaml` configuration file path for the model (here it's `TimesNet_cls.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/time_series_modules/time_series_classification.md

@@ -15,7 +15,7 @@ comments: true
 <tr>
 <th>模型名称</th><th>模型下载链接</th>
 <th>acc(%)</th>
-<th>模型存储大小(M)</th>
+<th>模型存储大小(M</th>
 <th>介绍</th>
 </tr>
 </thead>
@@ -453,7 +453,8 @@ python main.py -c paddlex/configs/modules/ts_classification/TimesNet_cls.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`TimesNet_cls.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 2 - 2
docs/module_usage/tutorials/time_series_modules/time_series_forecasting.en.md

@@ -509,8 +509,8 @@ You need to follow these steps:
 * Specify the `.yaml` configuration file path for the model (here it's `DLinear.yaml`,When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md)).
 * Set the mode to model training: `-o Global.mode=train`
 * Specify the training dataset path: `-o Global.dataset_dir`
-
-Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* Other related parameters can be set by modifying the `Global` and `Train` fields in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to train using the first two GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the [PaddleX TS Configuration Parameters Documentation](../../instructions/config_parameters_time_series.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/time_series_modules/time_series_forecasting.md

@@ -15,7 +15,7 @@ comments: true
 <th>模型名称</th><th>模型下载链接</th>
 <th>mse</th>
 <th>mae</th>
-<th>模型存储大小(M)</th>
+<th>模型存储大小(M</th>
 <th>介绍</th>
 </tr>
 </thead>
@@ -520,7 +520,8 @@ python main.py -c paddlex/configs/modules/ts_forecast/DLinear.yaml \
 * 指定模型的`.yaml` 配置文件路径(此处为`DLinear.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX时序任务模型配置文件参数说明](../../instructions/config_parameters_time_series.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/video_modules/video_classification.en.md

@@ -9,7 +9,6 @@ The Video Classification Module is a crucial component in a computer vision syst
 
 ## II. List of Supported Models
 
-
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
@@ -424,7 +423,9 @@ the following steps are required:
 
 * Specify the path of the model's `.yaml` configuration file (here it is `PP-TSMv2-LCNetV2_8frames_uniform.yaml`. When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
-* Specify the path of the training dataset: `-o Global.dataset_dir`. Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* Specify the path of the training dataset: `-o Global.dataset_dir`.
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the first 2 GPUs: `-o Global.device=gpu:0,1`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>

+ 2 - 1
docs/module_usage/tutorials/video_modules/video_classification.md

@@ -425,7 +425,8 @@ python main.py -c paddlex/configs/modules/video_classification/PP-TSMv2-LCNetV2_
 * 指定模型的`.yaml` 配置文件路径(此处为`PP-TSMv2-LCNetV2_8frames_uniform.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md))
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定前 2 卡 gpu 训练:`-o Global.device=gpu:0,1`;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>
 

+ 3 - 3
docs/module_usage/tutorials/video_modules/video_detection.en.md

@@ -11,7 +11,6 @@ The output of the video detection module includes bounding boxes and class label
 
 ## II. List of Supported Models
 
-
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
@@ -408,8 +407,9 @@ the following steps are required:
 
 * Specify the path of the model's `.yaml` configuration file (here it is `YOWO.yaml`. When training other models, you need to specify the corresponding configuration files. The relationship between the model and configuration files can be found in the [PaddleX Model List (CPU/GPU)](../../../support_list/models_list.en.md))
 * Specify the mode as model training: `-o Global.mode=train`
-* Specify the path of the training dataset: `-o Global.dataset_dir`. Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the second GPU: `-o Global.device=gpu:2`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
-
+* Specify the path of the training dataset: `-o Global.dataset_dir`.
+* Other related parameters can be set by modifying the fields under `Global` and `Train` in the `.yaml` configuration file, or adjusted by appending parameters in the command line. For example, to specify training on the second GPU: `-o Global.device=gpu:2`; to set the number of training epochs to 10: `-o Train.epochs_iters=10`. For more modifiable parameters and their detailed explanations, refer to the configuration file parameter instructions for the corresponding task module of the model [PaddleX Common Model Configuration File Parameters](../../instructions/config_parameters_common.en.md).
+* New Feature: Paddle 3.0 support CINN (Compiler Infrastructure for Neural Networks) to accelerate training speed when using GPU device. Please specify `-o Train.dy2st=True` to enable it.
 
 <details><summary>👉 <b>More Details (Click to Expand)</b></summary>
 

+ 3 - 2
docs/module_usage/tutorials/video_modules/video_detection.md

@@ -413,10 +413,11 @@ python main.py -c paddlex/configs/modules/video_detection/YOWO.yaml  \
 ```
 需要如下几步:
 
-* 指定模型的`.yaml` 配置文件路径(此处为`YOWO.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md)
+* 指定模型的`.yaml` 配置文件路径(此处为`YOWO.yaml`,训练其他模型时,需要的指定相应的配置文件,模型和配置的文件的对应关系,可以查阅[PaddleX模型列表(CPU/GPU)](../../../support_list/models_list.md)
 * 指定模式为模型训练:`-o Global.mode=train`
 * 指定训练数据集路径:`-o Global.dataset_dir`
-其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定第 2 卡 gpu 训练:`-o Global.device=gpu:2`,视频检测只支持单卡训练;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 其他相关参数均可通过修改`.yaml`配置文件中的`Global`和`Train`下的字段来进行设置,也可以通过在命令行中追加参数来进行调整。如指定第 2 卡 gpu 训练:`-o Global.device=gpu:2`,视频检测只支持单卡训练;设置训练轮次数为 10:`-o Train.epochs_iters=10`。更多可修改的参数及其详细解释,可以查阅模型对应任务模块的配置文件说明[PaddleX通用模型配置文件参数说明](../../instructions/config_parameters_common.md)。
+* 新特性:Paddle 3.0 版本支持了 CINN 神经网络编译器,在使用 GPU 设备训练时,不同模型有不同程度的训练加速效果。在 PaddleX 中训练模型时,可通过指定参数 `-o Train.dy2st=True` 开启。
 
 <details><summary>👉 <b>更多说明(点击展开)</b></summary>