matreche/modelarts-umn
1
0
Fork 0
forked from docs/modelarts
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Reimport with latest import script (#22)

Browse Source 
Reimport UMN with latest import script

fix various links

Reviewed-by: Beibei <None>
Reviewed-by: kucerakk <kucerakk@gmail.com>
This commit is contained in:
Artem Goncharov 2022-04-13 09:34:02 +02:00 committed by GitHub
parent c7a1163575
commit c34050e171
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
203 changed files with 1636 additions and 1466 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
umn/source/change_history.rst
View File

@ -1,10 +1,8 @@
.. _modelarts_04_0099:
Change History
==============
.. _modelarts040099enustopic0135264638enustopic0135264638table4331195115321:
=========== ===================================
Released On Description
=========== ===================================

8
umn/source/custom_images/creating_and_uploading_a_custom_image.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0085:
Creating and Uploading a Custom Image
=====================================
@ -9,7 +11,9 @@ ModelArts allows you to use custom images to create training jobs and import mod
Obtain the custom images used by ModelArts for model training and import from the SWR service management list. Upload the custom images you create to SWR.
- Specifications for custom images. For details about how to use a custom image for a training job, see `Specifications for Custom Images Used for Training Jobs <../custom_images/for_training_models/specifications_for_custom_images_used_for_training_jobs.html>`__. For details about how to use a custom image for model import, see `Specifications for Custom Images Used for Importing Models <../custom_images/for_importing_models/specifications_for_custom_images_used_for_importing_models.html>`__.
- Specifications for custom images. For details about how to use a custom image for a training job, see :ref:`Specifications for Custom Images Used for Training Jobs <modelarts_23_0217>`. For details about how to use a custom image for model import, see :ref:`Specifications for Custom Images Used for Importing Models <modelarts_23_0219>`.
.. _modelarts_23_0085__en-us_topic_0171858297_section125639162589:
.. _creating-and-uploading-a-custom-image-1:
@ -21,5 +25,3 @@ Creating and Uploading a Custom Image
#. Compile a Dockerfile based on your requirements to build a custom image. For details about how to efficiently compile a Dockerfile, see *SoftWare Repository for Container Best Practices*.
4. After customizing an image, upload the image to SWR by referring to "Uploading an Image Through a Docker Client" in *Software Repository for Container User Guide*.

14
umn/source/custom_images/for_importing_models/importing_a_model_using_a_custom_image.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0086:
Importing a Model Using a Custom Image
======================================
@ -6,13 +8,13 @@ After creating and uploading a custom image to SWR, you can use the image to imp
Prerequisites
-------------
- You have created a custom image package based on ModelArts specifications. For details about the specifications you need to comply with when using a custom image to import a model, see `Specifications for Custom Images Used for Importing Models <../../custom_images/for_importing_models/specifications_for_custom_images_used_for_importing_models.html>`__.
- You have uploaded the custom image to SWR. For details, see `Creating and Uploading a Custom Image <../../custom_images/creating_and_uploading_a_custom_image.html#creating-and-uploading-a-custom-image>`__.
- You have created a custom image package based on ModelArts specifications. For details about the specifications you need to comply with when using a custom image to import a model, see :ref:`Specifications for Custom Images Used for Importing Models <modelarts_23_0219>`.
- You have uploaded the custom image to SWR. For details, see :ref:`Creating and Uploading a Custom Image <modelarts_23_0085__en-us_topic_0171858297_section125639162589>`.
Importing a Model
-----------------
Set basic parameters for importing a model according to `Importing a Meta Model from a Container Image <../../model_management/importing_a_model/importing_a_meta_model_from_a_container_image.html>`__. When importing a model using a custom image, pay attention to the settings of **Meta Model Source** and **Configuration File**.
Set basic parameters for importing a model according to :ref:`Importing a Meta Model from a Container Image <modelarts_23_0206>`. When importing a model using a custom image, pay attention to the settings of **Meta Model Source** and **Configuration File**.
- **Meta Model Source**
@ -20,16 +22,14 @@ Set basic parameters for importing a model according to `Importing a Meta Model
- **Configuration File**
The model configuration file needs to be compiled independently. For details about how to compile the model configuration file, see `Specifications for Compiling the Model Configuration File <../../model_package_specifications/specifications_for_compiling_the_model_configuration_file.html>`__. For details about the configuration file examples of a custom image, see `Example of the Custom Image Model Configuration File <../../model_package_specifications/specifications_for_compiling_the_model_configuration_file.html#example-of-the-custom-image-model-configuration-file>`__. After editing the model configuration file based on the ModelArts specifications, upload it to OBS or use **Edit online** on the **Import Model** page.
The model configuration file needs to be compiled independently. For details about how to compile the model configuration file, see :ref:`Specifications for Compiling the Model Configuration File <modelarts_23_0092>`. For details about the configuration file examples of a custom image, see :ref:`Example of the Custom Image Model Configuration File <modelarts_23_0092__en-us_topic_0172466149_section9113122232018>`. After editing the model configuration file based on the ModelArts specifications, upload it to OBS or use **Edit online** on the **Import Model** page.
Deploying a Service
-------------------
After a model is successfully imported using a custom image, that is, the model status is normal, you can deploy the model as a service. On the **Models** page, click **Deploy** in the **Operation** column and select a service type, for example, **Real-time Service**.
You can deploy models as real-time or batch services based on the business logic of your custom image. The procedure for deploying a model imported using other methods is the same as that for deploying a model imported using a custom image. For details, see `Introduction to Model Deployment <../../model_deployment/introduction_to_model_deployment.html>`__.
You can deploy models as real-time or batch services based on the business logic of your custom image. The procedure for deploying a model imported using other methods is the same as that for deploying a model imported using a custom image. For details, see :ref:`Introduction to Model Deployment <modelarts_23_0058>`.
.. |image1| image:: /_static/images/en-us_image_0000001156920767.png

2
umn/source/custom_images/for_importing_models/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0218:
====================
For Importing Models
====================

8
umn/source/custom_images/for_importing_models/specifications_for_custom_images_used_for_importing_models.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0219:
Specifications for Custom Images Used for Importing Models
==========================================================
@ -40,9 +42,7 @@ Specifications for Custom Images Used for Model Management
{"health": "true"}
- Status code
.. _modelarts230219enustopic0212179953table19701134515351:
- Status code
.. table:: **Table 1** Status code
@ -65,5 +65,3 @@ Specifications for Custom Images Used for Model Management
- **Image dependencies**
To deploy a batch service, install component packages such as Python, JRE/JDK, and ZIP in the image.

20
umn/source/custom_images/for_training_models/creating_a_training_job_using_a_custom_image_gpu.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0087:
Creating a Training Job Using a Custom Image (GPU)
==================================================
@ -6,13 +8,13 @@ After creating and uploading a custom image to SWR, you can use the image to cre
Prerequisites
-------------
- You have created a custom image package based on ModelArts specifications. For details about the specifications you need to comply with when using a custom image to create training jobs, see `Specifications for Custom Images Used for Training Jobs <../../custom_images/for_training_models/specifications_for_custom_images_used_for_training_jobs.html>`__.
- You have uploaded the custom image to SWR. For details, see `Creating and Uploading a Custom Image <../../custom_images/creating_and_uploading_a_custom_image.html>`__.
- You have created a custom image package based on ModelArts specifications. For details about the specifications you need to comply with when using a custom image to create training jobs, see :ref:`Specifications for Custom Images Used for Training Jobs <modelarts_23_0217>`.
- You have uploaded the custom image to SWR. For details, see :ref:`Creating and Uploading a Custom Image <modelarts_23_0085>`.
Creating a Training Job
-----------------------
Log in to the ModelArts management console and create a training job according to `Creating a Training Job <../../training_management/creating_a_training_job/index.html>`__. When using a custom image to create a job, pay attention to the settings of **Algorithm Source**, **Environment Variable**, and **Resource Pool**.
Log in to the ModelArts management console and create a training job according to :ref:`Creating a Training Job <modelarts_23_0235>`. When using a custom image to create a job, pay attention to the settings of **Algorithm Source**, **Environment Variable**, and **Resource Pool**.
- **Algorithm Source**
@ -20,6 +22,8 @@ Log in to the ModelArts management console and create a training job according t
- **Image Path**: SWR URL after the image is uploaded to SWR
.. _modelarts_23_0087__en-us_topic_0171858299_fig1610311596365:
.. figure:: /_static/images/en-us_image_0000001156920769.png
:alt: **Figure 1** SWR image address
@ -50,11 +54,9 @@ Log in to the ModelArts management console and create a training job according t
- **Environment Variable**
After the container is started, besides the environment variables added by configuring **Environment Variable** during training job creation, `Table 1 <#modelarts230087enustopic0171858299table341782301619>`__ lists other environment variables to be loaded. You can determine whether to use these environment variables in your own Python training script, or run the **{python_file_parameter}** command to pass the required parameters.
After the container is started, besides the environment variables added by configuring **Environment Variable** during training job creation, :ref:`Table 1 <modelarts_23_0087__en-us_topic_0171858299_table341782301619>` lists other environment variables to be loaded. You can determine whether to use these environment variables in your own Python training script, or run the **{python_file_parameter}** command to pass the required parameters.
.. _modelarts230087enustopic0171858299table341782301619:
.. _modelarts_23_0087__en-us_topic_0171858299_table341782301619:
.. table:: **Table 1** Optional environment variables
@ -87,12 +89,10 @@ Log in to the ModelArts management console and create a training job according t
Running a Training Job Created Using a Custom Image
---------------------------------------------------
After a custom image is uploaded to SWR, ModelArts is authorized to obtain and run the image by default when you create a training job using the custom image. When a custom image is run for the first time, the image is checked first. For details about the check, see `Specifications for Custom Images Used for Training Jobs <../../custom_images/for_training_models/specifications_for_custom_images_used_for_training_jobs.html>`__. The check failure cause is outputted in the log, and you can modify the image based on the log.
After a custom image is uploaded to SWR, ModelArts is authorized to obtain and run the image by default when you create a training job using the custom image. When a custom image is run for the first time, the image is checked first. For details about the check, see :ref:`Specifications for Custom Images Used for Training Jobs <modelarts_23_0217>`. The check failure cause is outputted in the log, and you can modify the image based on the log.
After the image is checked, the backend starts the custom image container to run the training job. You can view the training status based on the log.
.. note::
After an image is reviewed, the image does not need to be reviewed again when being used to create training jobs again.

2
umn/source/custom_images/for_training_models/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0216:
===================
For Training Models
===================

34
umn/source/custom_images/for_training_models/specifications_for_custom_images_used_for_training_jobs.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0217:
Specifications for Custom Images Used for Training Jobs
=======================================================
@ -13,9 +15,11 @@ Specifications
- To ensure that the log content can be displayed normally, the logs must be standard output.
- The default user of a custom image must be the user whose UID is **1101**.
- Custom images can be developed based on basic ModelArts images. For details about the supported basic images, see `Overview of a Basic Image Package <#overview-of-a-basic-image-package>`__.
- Custom images can be developed based on basic ModelArts images. For details about the supported basic images, see :ref:`Overview of a Basic Image Package <modelarts_23_0217__en-us_topic_0212179951_section1126616610513>`.
- Currently, the ModelArts backend does not support the download of open source installation packages. You are advised to install the dependency packages required for training in the custom image.
.. _modelarts_23_0217__en-us_topic_0212179951_section1126616610513:
Overview of a Basic Image Package
---------------------------------
@ -39,8 +43,10 @@ After customizing an image, upload it to SWR. Make sure that you have created an
Obtain basic images based on chip requirements:
- `CPU-based Basic Images <#cpu-based-basic-images>`__
- `GPU-based Basic Images <#gpu-based-basic-images>`__
- :ref:`CPU-based Basic Images <modelarts_23_0217__en-us_topic_0212179951_section2357164275019>`
- :ref:`GPU-based Basic Images <modelarts_23_0217__en-us_topic_0212179951_section125281544151710>`
.. _modelarts_23_0217__en-us_topic_0212179951_section2357164275019:
CPU-based Basic Images
----------------------
@ -51,11 +57,9 @@ Address for obtaining a basic image
swr.<region>.xxx.com/modelarts-job-dev-image/custom-cpu-base:1.3
`Table 1 <#modelarts230217enustopic0212179951table42317014714>`__ and `Table 2 <#modelarts230217enustopic0212179951table624501372>`__ list the components and tools used by basic images.
:ref:`Table 1 <modelarts_23_0217__en-us_topic_0212179951_table42317014714>` and :ref:`Table 2 <modelarts_23_0217__en-us_topic_0212179951_table624501372>` list the components and tools used by basic images.
.. _modelarts230217enustopic0212179951table42317014714:
.. _modelarts_23_0217__en-us_topic_0212179951_table42317014714:
.. table:: **Table 1** Components
@ -65,9 +69,7 @@ Address for obtaining a basic image
| run_train.sh | Training boot script. You can download the code directory, run training commands, redirect training log output, and upload log files to OBS after training commands are executed. |
+--------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
.. _modelarts230217enustopic0212179951table624501372:
.. _modelarts_23_0217__en-us_topic_0212179951_table624501372:
.. table:: **Table 2** Tool list
@ -85,6 +87,8 @@ Address for obtaining a basic image
| dls-downloader.py | OBS download script. The **utils.sh** script depends on this script. |
+-----------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+
.. _modelarts_23_0217__en-us_topic_0212179951_section125281544151710:
GPU-based Basic Images
----------------------
@ -99,10 +103,6 @@ Addresses for obtaining a basic image
swr.<region>.xxx.com/modelarts-job-dev-image/custom-base-cuda10.1-cp36-ubuntu18.04-x86:1.1
swr.<region>.xxx.com/modelarts-job-dev-image/custom-base-cuda10.2-cp36-ubuntu18.04-x86:1.1
.. _modelarts230217enustopic0212179951table137851182312:
.. table:: **Table 3** Components
+--------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
@ -111,10 +111,6 @@ Addresses for obtaining a basic image
| run_train.sh | Training boot script. You can download the code directory, run training commands, redirect training log output, and upload log files to OBS after training commands are executed. |
+--------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
.. _modelarts230217enustopic0212179951table135271650237:
.. table:: **Table 4** Tool list
+-----------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+
@ -130,5 +126,3 @@ Addresses for obtaining a basic image
+-----------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+
| dls-downloader.py | OBS download script. The **utils.sh** script depends on this script. |
+-----------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+

2
umn/source/custom_images/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0083:
=============
Custom Images
=============

4
umn/source/custom_images/introduction_to_custom_images.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0084:
Introduction to Custom Images
=============================
@ -17,5 +19,3 @@ Application Scenarios of Custom Images
- **For Importing Models**
If you use an AI engine that is not supported by ModelArts to develop a model, you can create a custom image, import the image to ModelArts for unified management, and deploy the model as a service.

202
umn/source/data_management/creating_a_dataset.rst
View File

File diff suppressed because it is too large Load Diff

4
umn/source/data_management/deleting_a_dataset.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0021:
Deleting a Dataset
==================
@ -16,5 +18,3 @@ Procedure
.. note::
After a dataset is deleted, some functions such as dataset version management become unavailable. Exercise caution when performing this operation. However, the original data and labeling data of the dataset are still stored in OBS.

25
umn/source/data_management/exporting_data.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0214:
Exporting Data
==============
@ -21,11 +23,13 @@ Exporting Data to a New Dataset
.. note::
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to `4 <#modelarts230214enustopic0209632492li114071010139>`__.
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to :ref:`4 <modelarts_23_0214__en-us_topic_0209632492_li114071010139>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed.
#. On the dataset details page, select or filter data to be exported. Click **Export To** and choose **New Dataset** from the drop-down list.
#. .. _modelarts_23_0214__en-us_topic_0209632492_li114071010139:
On the dataset details page, select or filter data to be exported. Click **Export To** and choose **New Dataset** from the drop-down list.
#. In the displayed **Export to New Dataset** dialog box, enter the related information and click **OK**.
@ -48,11 +52,13 @@ Exporting Data to OBS
.. note::
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to `4 <#modelarts230214enustopic0209632492li2056103713438>`__.
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to :ref:`4 <modelarts_23_0214__en-us_topic_0209632492_li2056103713438>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed.
#. On the dataset details page, select or filter data to be exported. Click **Export To** and choose **OBS** from the drop-down list.
#. .. _modelarts_23_0214__en-us_topic_0209632492_li2056103713438:
On the dataset details page, select or filter data to be exported. Click **Export To** and choose **OBS** from the drop-down list.
#. In the displayed **Export to OBS** dialog box, enter the related information and click **OK**.
@ -68,14 +74,17 @@ Viewing the Task History
When you export data to a new dataset or OBS, you can view the export task details in the **View Task History** dialog box.
#. Log in to the ModelArts management console. In the left navigation pane, choose **Data Management** > **Datasets**. The **Datasets** page is displayed.
#. In the dataset list, select the dataset of the object detection or image classification type and click the dataset name to go to the **Dashboard** tab page of the dataset.
.. note::
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to `4 <#modelarts230214enustopic0209632492li19995141771413>`__.
For a dataset of the free format type, you can click the dataset name to directly access the dataset details page and go to :ref:`4 <modelarts_23_0214__en-us_topic_0209632492_li19995141771413>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed.
#. On the dataset details page, select or filter data to be exported. Click **Export To** and choose **View Task History** from the drop-down list.
#. .. _modelarts_23_0214__en-us_topic_0209632492_li19995141771413:
On the dataset details page, select or filter data to be exported. Click **Export To** and choose **View Task History** from the drop-down list.
#. In the **View Task History** dialog box, view the export task history of the current dataset. Information about **Task ID**, **Created**, **Type**, **Path**, **Total**, and **Status** is included.

110
umn/source/data_management/importing_data/import_operation.rst
View File

File diff suppressed because it is too large Load Diff

2
umn/source/data_management/importing_data/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0005:
==============
Importing Data
==============

22
umn/source/data_management/importing_data/specifications_for_importing_data_from_an_obs_directory.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0008:
Specifications for Importing Data from an OBS Directory
=======================================================
@ -9,6 +11,8 @@ Only the following types of dataset support the **OBS path** import mode: **Imag
To import data from an OBS directory, you must have the read permission on the OBS directory.
.. _modelarts_23_0008__en-us_topic_0170886816_section570816190577:
Image Classification
--------------------
@ -66,12 +70,14 @@ Image Classification
- Only images in JPG, JPEG, PNG, and BMP formats are supported. The size of a single image cannot exceed 5 MB, and the total size of all images uploaded at a time cannot exceed 8 MB.
.. _modelarts_23_0008__en-us_topic_0170886816_section1371122614572:
Object Detection
----------------
- The simple mode of object detection requires users store labeled objects and their label files (in one-to-one relationship with the labeled objects) in the same directory. For example, if the name of the labeled object file is **IMG_20180919_114745.jpg**, the name of the label file must be **IMG_20180919_114745.xml**.
The label files for object detection must be in PASCAL VOC format. For details about the format, see `Table 8 <../../data_management/importing_data/specifications_for_importing_the_manifest_file.html#modelarts230009enustopic0170886817table77167388472>`__.
The label files for object detection must be in PASCAL VOC format. For details about the format, see :ref:`Table 8 <modelarts_23_0009__en-us_topic_0170886817_table77167388472>`.
Example:
@ -134,12 +140,14 @@ Object Detection
- Only images in JPG, JPEG, PNG, and BMP formats are supported. The size of a single image cannot exceed 5 MB, and the total size of all images uploaded at a time cannot exceed 8 MB.
.. _modelarts_23_0008__en-us_topic_0170886816_section1363851815518:
Image Segmentation
------------------
- The simple mode of image segmentation requires users store labeled objects and their label files (in one-to-one relationship with the labeled objects) in the same directory. For example, if the name of the labeled object file is **IMG_20180919_114746.jpg**, the name of the label file must be **IMG_20180919_114746.xml**.
Fields **mask_source** and **mask_color** are added to the label file in PASCAL VOC format. For details about the format, see `Table 4 <../../data_management/importing_data/specifications_for_importing_the_manifest_file.html#modelarts230009enustopic0170886817table1516151991311>`__.
Fields **mask_source** and **mask_color** are added to the label file in PASCAL VOC format. For details about the format, see :ref:`Table 4 <modelarts_23_0009__en-us_topic_0170886817_table1516151991311>`.
Example:
@ -199,6 +207,8 @@ Image Segmentation
| 39 | </annotation> |
+-----------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------+
.. _modelarts_23_0008__en-us_topic_0170886816_section163641141195713:
Text Classification
-------------------
@ -249,6 +259,8 @@ Text classification supports two import modes.
│ COMMENTS _20180919_114945.txt
│ COMMENTS _20180919_114945_result.txt
.. _modelarts_23_0008__en-us_topic_0170886816_section1683314458578:
Sound Classification
--------------------
@ -269,6 +281,8 @@ Example:
2.wav
3.wav
.. _modelarts_23_0008__en-us_topic_0170886816_section1171862514918:
Table
-----
@ -277,7 +291,7 @@ You can import data from OBS.
Import description:
#. The prerequisite for successful import is that the schema of the data source must be the same as that specified during dataset creation. The schema indicates column names and types of a table. Once specified during dataset creation, the values cannot be changed.
#. If the data format is invalid, the data is set to null values. For details, see `Table 4 <../../data_management/creating_a_dataset.html#modelarts230004enustopic0170886809table1916832104917>`__.
#. If the data format is invalid, the data is set to null values. For details, see :ref:`Table 4 <modelarts_23_0004__en-us_topic_0170886809_table1916832104917>`.
#. When a CSV file is imported from OBS, the data type is not verified, but the number of columns must be the same as that in the schema of the dataset.
- From OBS
@ -291,5 +305,3 @@ Import description:
│ table_import_2.csv
│ table_import_3.csv
│ table_import_4.csv

294
umn/source/data_management/importing_data/specifications_for_importing_the_manifest_file.rst
View File

File diff suppressed because it is too large Load Diff

2
umn/source/data_management/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0002:
===============
Data Management
===============

86
umn/source/data_management/introduction_to_data_management.rst
View File

File diff suppressed because it is too large Load Diff

14
umn/source/data_management/labeling_data/image_classification.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0011:
Image Classification
====================
@ -15,7 +17,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -41,6 +43,8 @@ The following filter criteria are supported. You can set one or more filter crit
- **File Name** or **Path**: Filter files by file name or file storage path.
- **Labeled By**: Select the name of the user who performs the labeling operation.
.. _modelarts_23_0011__en-us_topic_0170889731_section888019266174:
Labeling Images (Manually)
--------------------------
@ -55,7 +59,7 @@ The dataset details page displays images on the **All**, **Labeled**, and **Unla
a. In the label adding area on the right, set the label in the **Label** text box.
Click the **Label** text box and select an existing label from the drop-down list. If the existing labels cannot meet the requirements, you can go to the page for `modifying the dataset <../../data_management/modifying_a_dataset.html>`__ and add labels.
Click the **Label** text box and select an existing label from the drop-down list. If the existing labels cannot meet the requirements, you can go to the page for :ref:`modifying the dataset <modelarts_23_0020>` and add labels.
b. Confirm the **Labels of Selected Image** information and click **OK**. The selected image is automatically moved to the **Labeled** tab page. On the **Unlabeled** and **All** tab pages, the labeling information is updated along with the labeling process, including the added label names and the number of images for each label.
@ -77,6 +81,8 @@ After labeling data, you can modify labeled data on the **Labeled** tab page.
Deleting a label: In the **File Labels** area, click the delete icon in the **Operation** column to delete the label. This operation deletes only the labels added to the selected image.
.. _modelarts_23_0011__en-us_topic_0170889731_en-us_topic_0170889731_fig171368141175:
.. figure:: /_static/images/en-us_image_0000001156921011.png
:alt: **Figure 1** Modifying a label
@ -90,6 +96,8 @@ After labeling data, you can modify labeled data on the **Labeled** tab page.
- Modifying a label: Click the editing icon in the **Operation** column. In the dialog box that is displayed, enter the new label name and click **OK**. After the modification, the images that have been added with the label use the new label name.
- Deleting a label: Click the deletion icon in the **Operation** column. In the displayed dialog box, select **Delete label**, **Delete label and images with only the label (Do not delete source files)**, or **Delete label and images with only the label (Delete source files)**, and click **OK**.
.. _modelarts_23_0011__en-us_topic_0170889731_en-us_topic_0170889731_fig19495403277:
.. figure:: /_static/images/en-us_image_0000001157080983.png
:alt: **Figure 2** Information about all labels
@ -125,5 +133,3 @@ If a tick is displayed in the upper left corner of an image, the image is select
.. note::
If you select **Delete source files**, images stored in the corresponding OBS directory will be deleted when you delete the selected images. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.

24
umn/source/data_management/labeling_data/image_segmentation.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0345:
Image Segmentation
==================
@ -20,7 +22,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -51,25 +53,25 @@ Manually Labeling Images
The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **All** tab page is displayed by default.
#. On the **Unlabeled** tab page, click an image. The system automatically directs you to the page for labeling the image. For details about how to use common buttons on this page, see `Table 2 <#modelarts230345enustopic0000001126398947table194471512463>`__.
#. On the **Unlabeled** tab page, click an image. The system automatically directs you to the page for labeling the image. For details about how to use common buttons on this page, see :ref:`Table 2 <modelarts_23_0345__en-us_topic_0000001126398947_table194471512463>`.
#. Select a labeling method.
On the labeling page, common `labeling methods <#modelarts230345enustopic0000001126398947table165201739119>`__ and `buttons <#modelarts230345enustopic0000001126398947table194471512463>`__ are provided in the toolbar. By default, polygon labeling is selected. Use polygon or point labeling as needed.
On the labeling page, common :ref:`labeling methods <modelarts_23_0345__en-us_topic_0000001126398947_table165201739119>` and :ref:`buttons <modelarts_23_0345__en-us_topic_0000001126398947_table194471512463>` are provided in the toolbar. By default, polygon labeling is selected. Use polygon or point labeling as needed.
.. note::
After you select a method to label the first image, the labeling method automatically applies to subsequent images.
.. _modelarts_23_0345__en-us_topic_0000001126398947_fig1362531203220:
.. figure:: /_static/images/en-us_image_0000001110920986.png
:alt: **Figure 1** Toolbar
**Figure 1** Toolbar
.. _modelarts230345enustopic0000001126398947table165201739119:
.. _modelarts_23_0345__en-us_topic_0000001126398947_table165201739119:
.. table:: **Table 1** Labeling methods
@ -81,9 +83,7 @@ The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **
| |image4| | Point labeling. Label the top, bottom, leftmost, and rightmost points on the object contour. The system will infer the outline of the object based on the labeled points. |
+----------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
.. _modelarts230345enustopic0000001126398947table194471512463:
.. _modelarts_23_0345__en-us_topic_0000001126398947_table194471512463:
.. table:: **Table 2** Toolbar buttons
@ -115,6 +115,8 @@ The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **
After labeling an image, click an image that has not been labeled in the image list below to label the new image.
.. _modelarts_23_0345__en-us_topic_0000001126398947_fig16575195124518:
.. figure:: /_static/images/en-us_image_0000001110761086.gif
:alt: **Figure 2** Labeling an object outline
@ -143,6 +145,8 @@ On the dataset details page, click the **Labeled** tab and then the image to be
After the labeling information is modified, click **Back to Data Labeling Preview** in the upper left part of the page to exit the labeling page. In the dialog box that is displayed, click **OK** to save the modification.
.. _modelarts_23_0345__en-us_topic_0000001126398947_en-us_topic_0170889732_fig16709173213107:
.. figure:: /_static/images/en-us_image_0000001156920955.gif
:alt: **Figure 3** Editing image labeling information
@ -179,8 +183,6 @@ If a tick is displayed in the upper left corner of an image, the image is select
If you select **Delete source files**, images stored in the corresponding OBS directory will be deleted when you delete the selected images. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001110920998.png
.. |image2| image:: /_static/images/en-us_image_0000001156920965.png

2
umn/source/data_management/labeling_data/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0010:
=============
Labeling Data
=============

12
umn/source/data_management/labeling_data/named_entity_recognition.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0014:
Named Entity Recognition
========================
@ -12,10 +14,12 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
.. _modelarts_23_0014__en-us_topic_0170889734_section888019266174:
Labeling Content
----------------
@ -32,6 +36,8 @@ Adding Labels
- Adding labels on the **Unlabeled** tab page: Click the plus sign (+) next to **Label Set**. On the **Add Label** page that is displayed, add a label name, select a label color, and click **OK**.
.. _modelarts_23_0014__en-us_topic_0170889734_fig162371842293:
.. figure:: /_static/images/en-us_image_0000001156921015.png
:alt: **Figure 1** Adding a named entity label (1)
@ -40,6 +46,8 @@ Adding Labels
- Adding labels on the **Labeled** tab page: Click the plus sign (+) next to **All Labels**. On the **Add Label** page that is displayed, add a label name, select a label color, and click **OK**.
.. _modelarts_23_0014__en-us_topic_0170889734_fig1418544013104:
.. figure:: /_static/images/en-us_image_0000001156921017.png
:alt: **Figure 2** Adding a named entity label (2)
@ -94,7 +102,5 @@ You can quickly delete the files you want to discard.
The background of the selected text is blue.
.. |image1| image:: /_static/images/en-us_image_0000001110761148.png

22
umn/source/data_management/labeling_data/object_detection.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0012:
Object Detection
================
@ -16,7 +18,7 @@ Labeling the Dataset
#. In the dataset list, click the dataset to be labeled based on the labeling type. The **Dashboard** tab page of the dataset is displayed.
By default, the **Dashboard** tab page of the current dataset version is displayed. To label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. To label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -42,12 +44,14 @@ The following filter criteria are supported. You can set one or more filter crit
- **File Name** or **Path**: Filter files by file name or file storage path.
- **Labeled By**: Select the name of the user who performs the labeling operation.
.. _modelarts_23_0012__en-us_topic_0170889732_section888019266174:
Labeling Images (Manually)
--------------------------
The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **All** tab page is displayed by default.
#. On the **Unlabeled** tab page, click an image. The image labeling page is displayed. For details about how to use common buttons on the **Labeled** tab page, see `Table 2 <#modelarts230012enustopic0170889732table194471512463>`__.
#. On the **Unlabeled** tab page, click an image. The image labeling page is displayed. For details about how to use common buttons on the **Labeled** tab page, see :ref:`Table 2 <modelarts_23_0012__en-us_topic_0170889732_table194471512463>`.
#. In the left tool bar, select a proper labeling shape. The default labeling shape is a rectangle. In this example, the rectangle is used for labeling.
@ -55,9 +59,7 @@ The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **
On the left of the page, multiple tools are provided for you to label images. However, you can use only one tool at a time.
.. _modelarts230012enustopic0170889732table165201739119:
.. _modelarts_23_0012__en-us_topic_0170889732_table165201739119:
.. table:: **Table 1** Supported bounding box
@ -85,9 +87,7 @@ The dataset details page provides the **Labeled** and **Unlabeled** tabs. The **
The selected image is automatically moved to the **Labeled** tab page. On the **Unlabeled** and **All** tab pages, the labeling information is updated along with the labeling process, including the added label names and the number of images for each label.
.. _modelarts230012enustopic0170889732table194471512463:
.. _modelarts_23_0012__en-us_topic_0170889732_table194471512463:
.. table:: **Table 2** Common icons on the labeling page
@ -133,6 +133,8 @@ After labeling data, you can modify labeled data on the **Labeled** tab page.
After deleting the label, click **Back to Data Labeling Preview** in the upper left corner of the page to exit the labeling page. In the dialog box that is displayed, save the modification. After all labels of an image are deleted, the image is displayed on the **Unlabeled** tab page.
.. _modelarts_23_0012__en-us_topic_0170889732_en-us_topic_0170889732_fig16709173213107:
.. figure:: /_static/images/en-us_image_0000001157080933.png
:alt: **Figure 1** Editing an object detection label
@ -146,6 +148,8 @@ After labeling data, you can modify labeled data on the **Labeled** tab page.
- Modifying a label: Click the edit icon in the **Operation** column. In the dialog box that is displayed, enter the new label name, select the new label color, and click **OK**. After the modification, the images that have been added with the label use the new label name.
- Deleting a label: Click the deletion icon in the **Operation** column to delete a label.
.. _modelarts_23_0012__en-us_topic_0170889732_en-us_topic_0170889732_fig19495403277:
.. figure:: /_static/images/en-us_image_0000001157080935.png
:alt: **Figure 2** All labels for object detection
@ -182,8 +186,6 @@ If a tick is displayed in the upper left corner of an image, the image is select
If you select **Delete source files**, images stored in the OBS directory will be deleted accordingly. This operation may affect other dataset versions or datasets using those files, for example, leading to an error in page display, training, or inference. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001156920971.png
.. |image2| image:: /_static/images/en-us_image_0000001156920969.png

8
umn/source/data_management/labeling_data/sound_classification.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0015:
Sound Classification
====================
@ -10,7 +12,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -68,6 +70,8 @@ After labeling data, you can modify labeled data on the **Labeled** tab page.
On the dataset details page, click the **Labeled** tab. The information about all labels is displayed on the right.
.. _modelarts_23_0015__en-us_topic_0170889735_fig19495403277:
.. figure:: /_static/images/en-us_image_0000001110761044.png
:alt: **Figure 1** Information about all labels
@ -105,7 +109,5 @@ If a tick is displayed in the upper right corner of an audio file, the audio fil
If you select **Delete source files**, audio files stored in the corresponding OBS directory will be deleted when you delete the selected audio files. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001157080893.png

8
umn/source/data_management/labeling_data/speech_labeling.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0016:
Speech Labeling
===============
@ -10,7 +12,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -32,6 +34,8 @@ The dataset details page displays the labeled and unlabeled audio files. The **U
#. After entering the content, click **OK** to complete the labeling. The audio file is automatically moved to the **Labeled** tab page.
.. _modelarts_23_0016__en-us_topic_0170889736_fig1525911501178:
.. figure:: /_static/images/en-us_image_0000001110920914.png
:alt: **Figure 1** Labeling an audio file
@ -76,7 +80,5 @@ On the **Unlabeled** or **Labeled** tab page, select the audio files to be delet
If you select **Delete source files**, audio files stored in the corresponding OBS directory will be deleted when you delete the selected audio files. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001110761012.png

8
umn/source/data_management/labeling_data/speech_paragraph_labeling.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0017:
Speech Paragraph Labeling
=========================
@ -10,7 +12,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -30,6 +32,8 @@ The dataset details page displays the labeled and unlabeled audio files. The **U
#. Select an audio segment based on the content being played, and enter the audio file label and content in the **Speech Content** text box.
.. _modelarts_23_0017__en-us_topic_0170889737_fig116336410274:
.. figure:: /_static/images/en-us_image_0000001157080965.png
:alt: **Figure 1** Labeling an audio file
@ -77,8 +81,6 @@ On the **Unlabeled** or **Labeled** tab page, select the audio files to be delet
If you select **Delete source files**, audio files stored in the corresponding OBS directory will be deleted when you delete the selected audio files. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001110761012.png
.. |image2| image:: /_static/images/en-us_image_0000001156920989.png

14
umn/source/data_management/labeling_data/text_classification.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0013:
Text Classification
===================
@ -15,10 +17,12 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
.. _modelarts_23_0013__en-us_topic_0170889733_section888019266174:
Labeling Content
----------------
@ -28,6 +32,8 @@ The dataset details page displays the labeled and unlabeled text files in the da
You can repeat this operation to select objects and add labels to the objects.
.. _modelarts_23_0013__en-us_topic_0170889733_fig127381972311:
.. figure:: /_static/images/en-us_image_0000001110760906.png
:alt: **Figure 1** Labeling for text classification
@ -41,6 +47,8 @@ Adding Labels
- Adding labels on the **Unlabeled** tab page: Click the plus sign (+) next to **Label Set**. On the **Add Label** page that is displayed, add a label name, select a label color, and click **OK**.
.. _modelarts_23_0013__en-us_topic_0170889733_fig162371842293:
.. figure:: /_static/images/en-us_image_0000001157080759.png
:alt: **Figure 2** Adding a label (1)
@ -49,6 +57,8 @@ Adding Labels
- Adding labels on the **Labeled** tab page: Click the plus sign (+) next to **All Labels**. On the **Add Label** page that is displayed, add a label name, select a label color, and click **OK**.
.. _modelarts_23_0013__en-us_topic_0170889733_fig1418544013104:
.. figure:: /_static/images/en-us_image_0000001110760912.png
:alt: **Figure 3** Adding a label (2)
@ -105,7 +115,5 @@ You can quickly delete the files you want to discard.
The background of the selected text is blue.
.. |image1| image:: /_static/images/en-us_image_0000001110760908.png

12
umn/source/data_management/labeling_data/text_triplet.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0211:
Text Triplet
============
@ -22,10 +24,12 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
.. _modelarts_23_0211__en-us_topic_0209128667_section888019266174:
Labeling Content
----------------
@ -33,6 +37,8 @@ The dataset details page displays the labeled and unlabeled text objects in the
#. On the **Unlabeled** tab page, the objects to be labeled are listed in the left pane. In the list, click a text object, select the corresponding text content on the right pane, and select an entity name from the displayed entity list to label the content.
.. _modelarts_23_0211__en-us_topic_0209128667_fig127381972311:
.. figure:: /_static/images/en-us_image_0000001110760968.png
:alt: **Figure 1** Labeling an entity
@ -41,6 +47,8 @@ The dataset details page displays the labeled and unlabeled text objects in the
#. After labeling multiple entities, click the source entity and target entity in sequence and select a relationship type from the displayed relationship list to label the relationship.
.. _modelarts_23_0211__en-us_topic_0209128667_fig16874184518477:
.. figure:: /_static/images/en-us_image_0000001110920874.png
:alt: **Figure 2** Labeling a relationship
@ -84,5 +92,3 @@ You can quickly delete the files you want to discard.
- On the **Labeled** tab page, select the text to be deleted and click **Delete**. Alternatively, you can tick **Select Images on Current Page** to select all text objects on the current page and click **Delete** in the upper left corner.
The background of the selected text is blue. If no text is selected on the page, the **Delete** button is unavailable.

20
umn/source/data_management/labeling_data/video_labeling.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0282:
Video Labeling
==============
@ -10,7 +12,7 @@ Starting Labeling
#. In the dataset list, select the dataset to be labeled based on the labeling type, and click the dataset name to go to the **Dashboard** tab page of the dataset.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see `Managing Dataset Versions <../../data_management/managing_dataset_versions.html>`__.
By default, the **Dashboard** tab page of the current dataset version is displayed. If you need to label the dataset of another version, click the **Versions** tab and then click **Set to Current Version** in the right pane. For details, see :ref:`Managing Dataset Versions <modelarts_23_0019>`.
#. On the **Dashboard** page of the dataset, click **Label** in the upper right corner. The dataset details page is displayed. By default, all data of the dataset is displayed on the dataset details page.
@ -30,9 +32,13 @@ On the dataset details page, both unlabeled and labeled video files in the datas
#. Play the video. When the video is played to the time point to be labeled, click the pause button in the progress bar to pause the video to a specific image.
#. In the left pane, select a bounding box. By default, a rectangular box is selected. Drag the mouse to select an object in the video image, enter a new label name in the displayed **Add Label** text box, select a label color, and click **Add** to label the object. Alternatively, select an existing label from the drop-down list and click **Add** to label the object. Label all objects in the image. Multiple labels can be added to an image.
#. .. _modelarts_23_0282__en-us_topic_0257844727_li993163014399:
The supported bounding boxes are the same as those supported by Object Detection. For details, see `Table 1 <../../data_management/labeling_data/object_detection.html#modelarts230012enustopic0170889732table165201739119>`__ in `Object Detection <../../data_management/labeling_data/object_detection.html>`__.
In the left pane, select a bounding box. By default, a rectangular box is selected. Drag the mouse to select an object in the video image, enter a new label name in the displayed **Add Label** text box, select a label color, and click **Add** to label the object. Alternatively, select an existing label from the drop-down list and click **Add** to label the object. Label all objects in the image. Multiple labels can be added to an image.
The supported bounding boxes are the same as those supported by Object Detection. For details, see :ref:`Table 1 <modelarts_23_0012__en-us_topic_0170889732_table165201739119>` in :ref:`Object Detection <modelarts_23_0012>`.
.. _modelarts_23_0282__en-us_topic_0257844727_fig87011122454:
.. figure:: /_static/images/en-us_image_0000001110761112.png
:alt: **Figure 1** Labeling video files
@ -40,10 +46,12 @@ On the dataset details page, both unlabeled and labeled video files in the datas
**Figure 1** Labeling video files
#. After the previous image is labeled, click the play button on the progress bar to resume the playback. Then, repeat `3 <#modelarts230282enustopic0257844727li993163014399>`__ to complete labeling on the entire video.
#. After the previous image is labeled, click the play button on the progress bar to resume the playback. Then, repeat :ref:`3 <modelarts_23_0282__en-us_topic_0257844727_li993163014399>` to complete labeling on the entire video.
The labeled time points of the current video are displayed on the right of the page.
.. _modelarts_23_0282__en-us_topic_0257844727_fig629913537509:
.. figure:: /_static/images/en-us_image_0000001156920985.png
:alt: **Figure 2** File labels
@ -64,6 +72,8 @@ On the **Labeled** tab page, click the target video file. In the **All Labels**
- Modifying a label: Click the edit icon on the right of a label to modify the label name.
- Deleting a label: Click the delete icon on the right of a label to delete the label. If you click the delete icon on the right of the image time, all labels on the image are deleted.
.. _modelarts_23_0282__en-us_topic_0257844727_fig338933705619:
.. figure:: /_static/images/en-us_image_0000001156920983.png
:alt: **Figure 3** Modifying labeled data
@ -83,7 +93,5 @@ If a tick is displayed in the upper left corner of a video file, the video file
If you select **Delete source files**, video files stored in the corresponding OBS directory will be deleted when you delete the selected video files. Deleting source files may affect other dataset versions or datasets using those files. As a result, the page display, training, or inference is abnormal. Deleted data cannot be recovered. Exercise caution when performing this operation.
.. |image1| image:: /_static/images/en-us_image_0000001110921012.png

8
umn/source/data_management/managing_dataset_versions.rst
View File

@ -1,9 +1,11 @@
.. _modelarts_23_0019:
Managing Dataset Versions
=========================
After labeling data, you can publish the dataset to multiple versions for management. For the published versions, you can view the dataset version updates, set the current version, and delete versions. For details about dataset versions, see `About Dataset Versions <../data_management/publishing_a_dataset.html#about-dataset-versions>`__.
After labeling data, you can publish the dataset to multiple versions for management. For the published versions, you can view the dataset version updates, set the current version, and delete versions. For details about dataset versions, see :ref:`About Dataset Versions <modelarts_23_0018__en-us_topic_0170886812_section38541340654>`.
For details about how to publish a new version, see `Publishing a Dataset <../data_management/publishing_a_dataset.html>`__.
For details about how to publish a new version, see :ref:`Publishing a Dataset <modelarts_23_0018>`.
Viewing Dataset Version Updates
-------------------------------
@ -35,5 +37,3 @@ Deleting a Dataset Version
.. note::
Deleting a dataset version does not remove the original data. Data and its labeling information are still stored in the OBS directory. However, if it is deleted, you cannot manage the dataset versions on the ModelArts management console. Exercise caution when performing this operation.

26
umn/source/data_management/modifying_a_dataset.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0020:
Modifying a Dataset
===================
@ -17,20 +19,18 @@ Modifying the Basic Information About a Dataset
Alternatively, you can click the dataset name to go to the **Dashboard** tab page of the dataset, and click **Modify** in the upper right corner.
#. Modify basic information about the dataset and then click **OK**. Refer to `Table 1 <#modelarts230020enustopic0170886811table151481125214>`__ for details.
#. Modify basic information about the dataset and then click **OK**. Refer to :ref:`Table 1 <modelarts_23_0020__en-us_topic_0170886811_table151481125214>` for details.
.. _modelarts230020enustopic0170886811table151481125214:
.. _modelarts_23_0020__en-us_topic_0170886811_table151481125214:
.. table:: **Table 1** Parameters
+-------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Parameter | Description |
+=============+==========================================================================================================================================================================================================================+
| Name | Enter the name of the dataset. A dataset name can contain only letters, digits, underscores (_), and hyphens (-). |
+-------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Description | Enter a brief description for the dataset. |
+-------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Label Set | The label set varies depending on the dataset type. For details about how to modify the label set, see the parameters of different dataset types in `Creating a Dataset <../data_management/creating_a_dataset.html>`__. |
+-------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
+-------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Parameter | Description |
+=============+====================================================================================================================================================================================================+
| Name | Enter the name of the dataset. A dataset name can contain only letters, digits, underscores (_), and hyphens (-). |
+-------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Description | Enter a brief description for the dataset. |
+-------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Label Set | The label set varies depending on the dataset type. For details about how to modify the label set, see the parameters of different dataset types in :ref:`Creating a Dataset <modelarts_23_0004>`. |
+-------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

10
umn/source/data_management/publishing_a_dataset.rst
View File

@ -1,8 +1,12 @@
.. _modelarts_23_0018:
Publishing a Dataset
====================
ModelArts distinguishes data of the same source according to versions labeled at different time, which facilitates the selection of dataset versions during subsequent model building and development. After labeling the data, you can publish the dataset to generate a new dataset version.
.. _modelarts_23_0018__en-us_topic_0170886812_section38541340654:
About Dataset Versions
----------------------
@ -23,9 +27,7 @@ Publishing a Dataset
Alternatively, you can click the dataset name to go to the **Dashboard** tab page of the dataset, and click **Publish** in the upper right corner.
#. In the displayed dialog box, set the parameters and click **OK**.
.. _modelarts230018enustopic0170886812table856411819131:
#. In the displayed dialog box, set the parameters and click **OK**.
.. table:: **Table 1** Parameters for publishing a dataset
@ -85,5 +87,3 @@ The following uses object detection as an example. If a manifest file is importe
|-- VersionMame2
...
|-- ...

2
umn/source/data_management/team_labeling/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0180:
=============
Team Labeling
=============

20
umn/source/data_management/team_labeling/introduction_to_team_labeling.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0181:
Introduction to Team Labeling
=============================
@ -12,13 +14,17 @@ How to Enable Team Labeling
- When creating a dataset, enable **Team Labeling** and select a team or task manager.
.. _modelarts_23_0181__en-us_topic_0186456616_fig19662182219716:
.. figure:: /_static/images/en-us_image_0000001157080899.png
:alt: **Figure 1** Enabling during dataset creation
**Figure 1** Enabling during dataset creation
- If team labeling is not enabled for a dataset that has been created, create a team labeling task to enable team labeling. For details about how to create a team labeling task, see `Creating Team Labeling Tasks <../../data_management/team_labeling/managing_team_labeling_tasks.html#creating-team-labeling-tasks>`__.
- If team labeling is not enabled for a dataset that has been created, create a team labeling task to enable team labeling. For details about how to create a team labeling task, see :ref:`Creating Team Labeling Tasks <modelarts_23_0210__en-us_topic_0209053802_section72262410214>`.
.. _modelarts_23_0181__en-us_topic_0186456616_fig1943110322817:
.. figure:: /_static/images/en-us_image_0000001156921451.png
:alt: **Figure 2** Creating a team labeling task in a dataset list
@ -26,12 +32,16 @@ How to Enable Team Labeling
**Figure 2** Creating a team labeling task in a dataset list
.. _modelarts_23_0181__en-us_topic_0186456616_fig183348421489:
.. figure:: /_static/images/en-us_image_0000001110761582.png
:alt: **Figure 3** Creating a team labeling task
**Figure 3** Creating a team labeling task
.. _modelarts_23_0181__en-us_topic_0186456616_fig1542082785810:
.. figure:: /_static/images/en-us_image_0000001110761054.png
:alt: **Figure 4** Creating a team labeling task on the dataset details page
@ -41,8 +51,6 @@ How to Enable Team Labeling
Operations Related to Team Labeling
-----------------------------------
- `Team Management <../../data_management/team_labeling/team_management.html>`__
- `Member Management <../../data_management/team_labeling/member_management.html>`__
- `Managing Team Labeling Tasks <../../data_management/team_labeling/managing_team_labeling_tasks.html>`__
- :ref:`Team Management <modelarts_23_0182>`
- :ref:`Member Management <modelarts_23_0183>`
- :ref:`Managing Team Labeling Tasks <modelarts_23_0210>`

24
umn/source/data_management/team_labeling/managing_team_labeling_tasks.rst
View File

@ -1,8 +1,12 @@
.. _modelarts_23_0210:
Managing Team Labeling Tasks
============================
For datasets with team labeling enabled, you can create team labeling tasks and assign the labeling tasks to different teams so that team members can complete the labeling tasks together. During data labeling, members can initiate acceptance, continue acceptance, and view acceptance reports.
.. _modelarts_23_0210__en-us_topic_0209053802_section72262410214:
Creating Team Labeling Tasks
----------------------------
@ -19,7 +23,7 @@ You can also create a team marking task and assign it to different members in th
- **Type**: Select a task type, **Team** or **Task Manager**.
- **Select Team**: If **Type** is set to **Team**, you need to select a team and members for labeling. The **Select Team** drop-down list lists the labeling teams and members created by the current account. For details about team management, see `Introduction to Team Labeling <../../data_management/team_labeling/introduction_to_team_labeling.html>`__.
- **Select Team**: If **Type** is set to **Team**, you need to select a team and members for labeling. The **Select Team** drop-down list lists the labeling teams and members created by the current account. For details about team management, see :ref:`Introduction to Team Labeling <modelarts_23_0181>`.
- **Select Task Manager**: If **Type** is set to **Task Manager**, you need to select one **Team Manager** member from all teams as the task manager.
@ -39,16 +43,18 @@ After a labeling task is created, the team member to which the task is assigned
In the email details, click the labeling task link and use your email address and initial password to log in to the labeling platform. After login, change the password. After logging in to the labeling platform, you can view the assigned labeling task and click the task name to go to the labeling page. The labeling method varies depending on the dataset type. For details, see the following:
- `Image Classification <../../data_management/labeling_data/image_classification.html#labeling-images-(manually)>`__
- `Object Detection <../../data_management/labeling_data/object_detection.html#labeling-images-(manually)>`__
- `Text Classification <../../data_management/labeling_data/text_classification.html#labeling-content>`__
- `Named Entity Recognition <../../data_management/labeling_data/named_entity_recognition.html#labeling-content>`__
- `Text Triplet <../../data_management/labeling_data/text_triplet.html#labeling-content>`__
- :ref:`Image Classification <modelarts_23_0011__en-us_topic_0170889731_section888019266174>`
- :ref:`Object Detection <modelarts_23_0012__en-us_topic_0170889732_section888019266174>`
- :ref:`Text Classification <modelarts_23_0013__en-us_topic_0170889733_section888019266174>`
- :ref:`Named Entity Recognition <modelarts_23_0014__en-us_topic_0170889734_section888019266174>`
- :ref:`Text Triplet <modelarts_23_0211__en-us_topic_0209128667_section888019266174>`
On the labeling platform, each member can view the images that are not labeled, to be corrected, rejected, to be reviewed, approved, and accepted. Pay attention to the images rejected by the administrator and the images to be corrected.
If the Reviewer role is assigned for a team labeling task, the labeling result needs to be reviewed. After the labeling result is reviewed, it is submitted to the administrator for acceptance.
.. _modelarts_23_0210__en-us_topic_0209053802_fig13465256141515:
.. figure:: /_static/images/en-us_image_0000001110760934.png
:alt: **Figure 1** Labeling platform
@ -86,10 +92,6 @@ Task Acceptance (Administrator)
Once the labeled data is accepted, team members cannot modify the labeling information. Only the dataset creator can modify the labeling information.
.. _modelarts230210enustopic0209053802table1372918217370:
.. table:: **Table 1** Parameters for finishing acceptance
+-----------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
@ -116,5 +118,3 @@ Deleting a Labeling Task
------------------------
On the **Labeling Progress** tab page, click **Delete** in the row where a labeling task to be deleted. After a task is deleted, the labeling details that are not accepted will be lost. Exercise caution when performing this operation. However, the original data in the dataset and the labeled data that has been accepted are still stored in the corresponding OBS bucket.

10
umn/source/data_management/team_labeling/member_management.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0183:
Member Management
=================
@ -5,6 +7,8 @@ There is no member in a new team. You need to add members who will participate i
A maximum of 100 members can be added to a team. If there are more than 100 members, add them to different teams for better management.
.. _modelarts_23_0183__en-us_topic_0186456618_section060323818470:
Adding a Member
---------------
@ -20,12 +24,16 @@ Adding a Member
Possible values of **Role** are **Labeler**, **Reviewer**, and **Team Manager**. Only one **Team Manager** can be set.
.. _modelarts_23_0183__en-us_topic_0186456618_fig2095294217492:
.. figure:: /_static/images/en-us_image_0000001156920939.png
:alt: **Figure 1** Adding a member
**Figure 1** Adding a member
.. _modelarts_23_0183__en-us_topic_0186456618_fig2953352181118:
.. figure:: /_static/images/en-us_image_0000001157081267.png
:alt: **Figure 2** Adding a member
@ -57,5 +65,3 @@ Deleting Members
- **Batch Deletion**
In the **Team Details** area, select members to be deleted and click **Delete**. In the dialog box that is displayed, click **OK**.

10
umn/source/data_management/team_labeling/team_management.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0182:
Team Management
===============
@ -7,7 +9,9 @@ Background
----------
- An account can have a maximum of 10 teams.
- An account must have at least one team to enable team labeling for datasets. If the account has no team, add a team by referring to `Adding a Team <#adding-a-team>`__.
- An account must have at least one team to enable team labeling for datasets. If the account has no team, add a team by referring to :ref:`Adding a Team <modelarts_23_0182__en-us_topic_0186456617_section165361815383>`.
.. _modelarts_23_0182__en-us_topic_0186456617_section165361815383:
Adding a Team
-------------
@ -18,7 +22,7 @@ Adding a Team
#. In the displayed **Add Team** dialog box, enter a team name and description and click **OK**. The labeling team is added.
The new team is displayed on the **Labeling Teams** page. You can view team details in the right pane. There is no member in the new team. Add members to the new team by referring to `Adding a Member <../../data_management/team_labeling/member_management.html#adding-a-member>`__.
The new team is displayed on the **Labeling Teams** page. You can view team details in the right pane. There is no member in the new team. Add members to the new team by referring to :ref:`Adding a Member <modelarts_23_0183__en-us_topic_0186456618_section060323818470>`.
Deleting a Team
---------------
@ -26,5 +30,3 @@ Deleting a Team
You can delete a team that is no longer used.
On the **Labeling Teams** page, select the target team and click **Delete**. In the dialog box that is displayed, click **OK**.

2
umn/source/devenviron_notebook/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0032:
=====================
DevEnviron (Notebook)
=====================

12
umn/source/devenviron_notebook/introduction_to_notebook.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0033:
Introduction to Notebook
========================
@ -6,6 +8,8 @@ ModelArts integrates the open-source Jupyter Notebook and JupyterLab to provide
- Jupyter Notebook is an interactive notebook. For details about how to perform operations on Jupyter Notebook, see `Jupyter Notebook Documentation <https://jupyter.org/documentation>`__.
- JupyterLab is an interactive development environment. It is a next-generation product of Jupyter Notebook. JupyterLab enables you to compile notebooks, operate terminals, edit MarkDown text, open interaction modes, and view CSV files and images. For details about how to perform operations on JupyterLab, see `JupyterLab Documentation <https://jupyterlab.readthedocs.io/en/stable/>`__.
.. _modelarts_23_0033__en-us_topic_0162690357_section191109611479:
Supported AI Engines
--------------------
@ -15,10 +19,6 @@ Each development environment supports multiple AI engines that run independently
- Each ModelArts notebook instance can use all supported engines.
.. _modelarts230033enustopic0162690357table13949522712:
.. table:: **Table 1** AI engines
+------------------------------------------+--------------------------------+----------------+
@ -48,7 +48,7 @@ Constraints
- For security purposes, the root permission is not granted to the notebook instances integrated in ModelArts. You can use the non-privileged user **jovyan** or **ma-user** (using **Multi-Engine**) to perform operations. Therefore, you cannot use **apt-get** to install the OS software.
- Notebook instances support only standalone training under the current AI engine framework. If you need to use distributed training, use ModelArts training jobs and specify multiple nodes in the resource pool.
- ModelArts DevEnviron does not support apt-get. You can use a `custom image <../custom_images/introduction_to_custom_images.html>`__ to train a model.
- ModelArts DevEnviron does not support apt-get. You can use a :ref:`custom image <modelarts_23_0084>` to train a model.
- Notebook instances do not support GUI-related libraries, such as PyQt.
- Notebook instances created using Ascend specifications cannot be attached to EVS disks.
- Notebook instances cannot be connected to DWS and database services.
@ -56,5 +56,3 @@ Constraints
- DevEnviron does not support TensorBoard. Use the visualization job function under **Training Jobs**.
- After a notebook instance is created, you cannot modify its specifications. For example, you cannot change the CPU specifications to GPU specifications or change the work environment. Therefore, select the specifications required by the service when creating a notebook instance, or save your code and data to OBS in a timely manner during development so that you can quickly upload the code and data to a new notebook instance.
- If the code output is still displayed after you close the page and open it again, use Terminal.

84
umn/source/devenviron_notebook/managing_notebook_instances/creating_a_notebook_instance.rst
View File

File diff suppressed because it is too large Load Diff

4
umn/source/devenviron_notebook/managing_notebook_instances/deleting_a_notebook_instance.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0042:
Deleting a Notebook Instance
============================
@ -9,5 +11,3 @@ You can delete notebook instances that are no longer used to release resources.
.. note::
Deleted notebook instances cannot be recovered. Therefore, exercise caution when performing this operation. However, the files created in notebook instances are still stored in OBS specified during creation of the notebook instances.

2
umn/source/devenviron_notebook/managing_notebook_instances/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0111:
===========================
Managing Notebook Instances
===========================

8
umn/source/devenviron_notebook/managing_notebook_instances/opening_a_notebook_instance.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0325:
Opening a Notebook Instance
===========================
@ -19,7 +21,5 @@ Code Development
ModelArts provides two environments for code development: Jupyter Notebook and JupyterLab.
- `Jupyter Notebook <../../devenviron_notebook/using_jupyter_notebook/introduction_to_jupyter_notebook.html>`__: a web-based application for interactive computing. It can be applied to full-process computing: development, documentation, running code, and presenting results.
- `JupyterLab <../../devenviron_notebook/using_jupyterlab/introduction_to_jupyterlab_and_common_operations.html>`__: an interactive development environment. It is a next-generation product of Jupyter Notebook. JupyterLab enables you to compile notebooks, operate terminals, edit MarkDown text, open interaction modes, and view CSV files and images.
- :ref:`Jupyter Notebook <modelarts_23_0326>`: a web-based application for interactive computing. It can be applied to full-process computing: development, documentation, running code, and presenting results.
- :ref:`JupyterLab <modelarts_23_0209>`: an interactive development environment. It is a next-generation product of Jupyter Notebook. JupyterLab enables you to compile notebooks, operate terminals, edit MarkDown text, open interaction modes, and view CSV files and images.

4
umn/source/devenviron_notebook/managing_notebook_instances/starting_or_stopping_a_notebook_instance.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0041:
Starting or Stopping a Notebook Instance
========================================
@ -7,5 +9,3 @@ Log in to the ModelArts management console. In the left navigation pane, choose
- To stop a notebook instance, locate the row where the notebook instance resides and click **Stop** in the **Operation** column. Only notebook instances in the **Running** state can be stopped.
- To start a notebook instance, locate the row where the notebook instance resides and click **Start** in the **Operation** column. Only notebook instances in the **Stopped** state can be started.

20
umn/source/devenviron_notebook/using_jupyter_notebook/common_operations_on_jupyter_notebook.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0120:
Common Operations on Jupyter Notebook
=====================================
@ -10,6 +12,8 @@ In the notebook instance list, locate the row where the target notebook instance
Two tab pages are available on the **Jupyter Notebook** page: **Files** and **Running**.
.. _modelarts_23_0120__en-us_topic_0188347008_fig13203124195913:
.. figure:: /_static/images/en-us_image_0000001110761034.png
:alt: **Figure 1** Jupyter Notebook
@ -21,6 +25,8 @@ Selecting Different AI Engines to Create Files
Open a notebook instance and go to the **Jupyter Notebook** page. On the **Files** tab page, click **New** in the upper right corner, select the required AI engine, and create a file for encoding.
.. _modelarts_23_0120__en-us_topic_0188347008_fig8224175513165:
.. figure:: /_static/images/en-us_image_0000001157080885.png
:alt: **Figure 2** Selecting different AI engines
@ -32,6 +38,8 @@ Uploading a File
Open a notebook instance and go to the **Jupyter Notebook** page. On the **Files** tab page, click **Upload** in the upper right corner to select a file from the local PC and upload it.
.. _modelarts_23_0120__en-us_topic_0188347008_fig89015882019:
.. figure:: /_static/images/en-us_image_0000001110920940.png
:alt: **Figure 3** Uploading a file
@ -43,16 +51,14 @@ Compiling a File
After a file is created, click the file name to go to the file compilation page.
.. _modelarts_23_0120__en-us_topic_0188347008_fig65191088443:
.. figure:: /_static/images/en-us_image_0000001110920938.png
:alt: **Figure 4** Compiling a file
**Figure 4** Compiling a file
.. _modelarts230120enustopic0188347008table9727162374411:
.. table:: **Table 1** Introduction to the file compilation page
+-----------------------+------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
@ -71,7 +77,7 @@ After a file is created, click the file name to go to the file compilation page.
| | | - Raw NBConvert: conversion tool. |
| | | - Heading: Quickly add a MarkDown title. |
+-----------------------+------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| 4 | AI engine and Python version | Displays the AI engine and Python version corresponding to the current file. For details about all AI engines and Python versions supported by ModelArts, see `Supported AI Engines <../../devenviron_notebook/introduction_to_notebook.html#supported-ai-engines>`__. |
| 4 | AI engine and Python version | Displays the AI engine and Python version corresponding to the current file. For details about all AI engines and Python versions supported by ModelArts, see :ref:`Supported AI Engines <modelarts_23_0033__en-us_topic_0162690357_section191109611479>`. |
+-----------------------+------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| 5 | Code cell | Each cell has two modes: command mode and editing mode. |
| | | |
@ -89,13 +95,13 @@ To delete a file or folder from Jupyter Notebook, select the file or folder in t
After the file or folder is deleted, click the **Refresh** button in the upper right corner to refresh the Jupyter page and clear the cache.
.. _modelarts_23_0120__en-us_topic_0188347008_fig10721575216:
.. figure:: /_static/images/en-us_image_0000001110761038.png
:alt: **Figure 5** Jupyter page
**Figure 5** Jupyter page
.. |image1| image:: /_static/images/en-us_image_0000001110920936.png

2
umn/source/devenviron_notebook/using_jupyter_notebook/configuring_the_jupyter_notebook_environment/index.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0327:
============================================
Configuring the Jupyter Notebook Environment
============================================

4
umn/source/devenviron_notebook/using_jupyter_notebook/configuring_the_jupyter_notebook_environment/installing_external_libraries_and_kernels_in_notebook_instances.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0040:
Installing External Libraries and Kernels in Notebook Instances
===============================================================
@ -46,5 +48,3 @@ Assume that you want to install Shapely from the terminal of a notebook instance
os.system('pip install Shapely')
import Shapely

6
umn/source/devenviron_notebook/using_jupyter_notebook/configuring_the_jupyter_notebook_environment/switching_the_cuda_version_on_the_terminal_page_of_a_gpu-based_notebook_instance.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0280:
Switching the CUDA Version on the Terminal Page of a GPU-based Notebook Instance
================================================================================
@ -21,10 +23,10 @@ CPU-based notebook instances do not use CUDA. Therefore, the following operation
sudo ln -snf /usr/local/cuda-10.0 cuda
.. _modelarts_23_0280__en-us_topic_0245881876_fig9219163419370:
.. figure:: /_static/images/en-us_image_0000001156920929.png
:alt: **Figure 1** Example of switching the CUDA version
**Figure 1** Example of switching the CUDA version

17
umn/source/devenviron_notebook/using_jupyter_notebook/configuring_the_jupyter_notebook_environment/using_the_notebook_terminal_function.rst
View File

@ -1,3 +1,5 @@
.. _modelarts_23_0117:
Using the Notebook Terminal Function
====================================
@ -10,6 +12,8 @@ Enabling the Notebook Terminal Function
#. On the **Files** tab page of the Jupyter page, click **New** and select **Terminal**. The **Terminal** page is displayed.
.. _modelarts_23_0117__en-us_topic_0190535990_fig98166612210:
.. figure:: /_static/images/en-us_image_0000001110920980.png
:alt: **Figure 1** Going to the Terminal page
@ -25,15 +29,10 @@ You can switch to another AI engine environment in the terminal environment of J
#. On the **Files** tab page of the Jupyter page, click **New** and select **Terminal**. The **Terminal** page is displayed.
#.
#. .. _modelarts_23_0117__en-us_topic_0190535990_fig161667313101:
.. container::
.. figure:: /_static/images/en-us_image_0000001110761076.png
:alt: **Figure 2** Output after command execution
.. figure:: /_static/images/en-us_image_0000001110761076.png
:alt: **Figure 2** Output after command execution
**Figure 2** Output after command execution
**Figure 2** Output after command execution

Some files were not shown because too many files have changed in this diff Show More