ModelArts 2021430 (GA) UMN 25072022 provided by R&D (third review)

Reviewed-by: gtema <artem.goncharov@gmail.com>
Co-authored-by: Jiang, Beibei <beibei.jiang@t-systems.com>
Co-committed-by: Jiang, Beibei <beibei.jiang@t-systems.com>

docs/modelarts/umn/modelarts_01_0001.html

@@ -4,6 +4,7 @@
 <div id="body8662426"><p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p1940111211504">ModelArts is a one-stop development platform for AI developers. With distributed training, automated model building, and model deployment, ModelArts helps AI developers quickly build models and efficiently manage the AI development lifecycle.</p>
 <p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p614433422614">ModelArts covers all stages of AI development, including data processing and model training and deployment. The underlying technologies of ModelArts support various heterogeneous computing resources, allowing developers to flexibly select and use resources. In addition, ModelArts supports popular open-source AI development frameworks such as TensorFlow and MXNet. Developers can also use self-developed algorithm frameworks to match their usage habits.</p>
 <p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p9658155902317">ModelArts aims to simplify AI development.</p>
+<p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p5701427143413"></p>
 <div class="section" id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_section18963850154919"><h4 class="sectiontitle">Product Architecture</h4><p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p2384838145615">ModelArts supports the entire development process, including data processing, and model training, management, and deployment.</p>
 <p id="modelarts_01_0001__en-us_topic_0284258732_en-us_topic_0129423679_p27301648182614">ModelArts supports various AI application scenarios, such as image classification and object detection.</p>
 </div>

docs/modelarts/umn/modelarts_01_0003.html

@@ -9,7 +9,6 @@
 </li><li id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_li1670516523416"><strong id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_b1193917524720">Multi-scenario deployment</strong><p id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_p159061027164218">Deploys models in various production environments, and supports real-time and batch inference.</p>
 </li><li id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_li18990194124211"><strong id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_b497484717306">Auto learning</strong><p id="modelarts_01_0003__en-us_topic_0284258733_en-us_topic_0129439806_p19708202220426">Enables model building without coding and supports image classification, object detection, and predictive analytics.</p>
 </li></ul>
-
 </div>
 <div>
 <div class="familylinks">

docs/modelarts/umn/modelarts_01_0006.html

@@ -1,11 +1,13 @@
 <a name="modelarts_01_0006"></a><a name="modelarts_01_0006"></a>
 
 <h1 class="topictitle1">Related Services</h1>
-<div id="body8662426"><div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section299615511534"><h4 class="sectiontitle">OBS</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p493914509919">ModelArts uses Object Storage Service (OBS) to store data and model backups and snapshots. OBS provides secure, reliable, low-cost storage. For more details, see <em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i3104928151115"></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i52511656161117"></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i1131122811129"><a href="https://docs.otc.t-systems.com/en-us/obs/index.html" target="_blank" rel="noopener noreferrer">Object Storage Service Console Function Overview</a></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i1919110719131"></em>.</p>
+<div id="body8662426"><div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section159941955105319"><h4 class="sectiontitle">IAM</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p11840154165516">ModelArts uses Identity and Access Management (IAM) for authentication and authorization. For more information about IAM, see <em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i569615374460"><a href="https://docs.otc.t-systems.com/en-us/iam/index.html" target="_blank" rel="noopener noreferrer">Identity and Access Management User Guide</a></em>.</p>
 </div>
-<div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section499755585318"><h4 class="sectiontitle">CCE</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p152905751010">ModelArts uses Cloud Container Engine (CCE) to deploy models as real-time services. CCE enables high concurrency and provides elastic scaling. For more information about CCE, see <u id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_u8895194525010"></u><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i225319514504"></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i179065818512"><a href="https://docs.otc.t-systems.com/en-us/cce/index.html" target="_blank" rel="noopener noreferrer">Cloud Container Engine User Guide</a></em>.</p>
+<div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section299615511534"><h4 class="sectiontitle">OBS</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p493914509919">ModelArts uses Object Storage Service (OBS) to store data and model backups and snapshots. OBS provides secure, reliable, low-cost storage. For more details, see <em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i1131122811129"><a href="https://docs.otc.t-systems.com/en-us/obs/index.html" target="_blank" rel="noopener noreferrer">Object Storage Service Console Function Overview</a></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i1919110719131"></em>.</p>
 </div>
-<div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section872515283144"><h4 class="sectiontitle">SWR</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p518511154151">To use an AI framework that is not supported by ModelArts, use SoftWare Repository for Container (SWR) to customize an image and import the image to ModelArts for training or inference. For more details, see <em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i1715418119248"></em><em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i11296534142412"></em>.</p>
+<div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section499755585318"><h4 class="sectiontitle">CCE</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p152905751010">ModelArts uses Cloud Container Engine (CCE) to deploy models as real-time services. CCE enables high concurrency and provides elastic scaling. For more information about CCE, see <em id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_i179065818512"><a href="https://docs.otc.t-systems.com/en-us/cce/index.html" target="_blank" rel="noopener noreferrer">Cloud Container Engine User Guide</a></em>.</p>
+</div>
+<div class="section" id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_section872515283144"><h4 class="sectiontitle">SWR</h4><p id="modelarts_01_0006__en-us_topic_0284258744_en-us_topic_0129439809_p518511154151">To use an AI framework that is not supported by ModelArts, use SoftWare Repository for Container (SWR) to customize an image and import the image to ModelArts for training or inference. For more details, see <em id="modelarts_01_0006__i11943153718013"><a href="https://docs.otc.t-systems.com/en-us/swr/index.html" target="_blank" rel="noopener noreferrer">SoftWare Repository for Container User Guide</a></em></p>
 </div>
 </div>
 <div>

docs/modelarts/umn/modelarts_01_0012.html

@@ -1,8 +1,9 @@
 <a name="modelarts_01_0012"></a><a name="modelarts_01_0012"></a>
 
 <h1 class="topictitle1">Data Management</h1>
-<div id="body8662426"><p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p179941317142417">During AI development, massive volumes of data need to be processed, and data preparation and labeling usually take more than half of the development time. ModelArts data management provides an efficient data management and labeling framework. It supports various data types such as image, text, audio, and video in a range of labeling scenarios such as image classification, object detection, speech paragraph labeling, and text classification. ModelArts data management can be used in AI projects of computer vision, natural language processing, and audio and video analysis. In addition, it provides functions such as data filtering, data analysis, team labeling, and version management for full-process data labeling. </p>
+<div id="body8662426"><p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p179941317142417">During AI development, massive volumes of data need to be processed, and data preparation and labeling usually take more than half of the development time. ModelArts data management provides an efficient data management and labeling framework. It supports various data types such as image, text, audio, and video in a range of labeling scenarios such as image classification, object detection, speech paragraph labeling, and text classification. ModelArts data management can be used in AI projects of computer vision, natural language processing, and audio and video analysis. In addition, it provides functions such as data filtering, data analysis, team labeling, and version management for full-process data labeling.</p>
 <div class="fignone" id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_fig156962483235"><span class="figcap"><b>Figure 1 </b>Data labeling process</span><br><span><img id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_image416983541715" src="en-us_image_0000001157080947.png"></span></div>
+<p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p9139454194317"></p>
 <p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p13303163511237">Team labeling enables multiple members to label a dataset, improving labeling efficiency. ModelArts allows project-based management for labeling by individual developers, small-scale labeling by small teams, and large-scale labeling by professional teams.</p>
 <p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p0303935202310">For large-scale team labeling, ModelArts provides team management, personnel management, and data management to implement the entire process, from project creation, allocation, management, labeling, to acceptance. For small-scale labeling by individuals and small teams, ModelArts provides an easy-to-use labeling tool to minimize project management costs.</p>
 <p id="modelarts_01_0012__en-us_topic_0284258738_en-us_topic_0168462753_p10303153522316">In addition, the labeling platform ensures data security. User data is used only within the authorized scope. The labeling object allocation policy ensures user data privacy and implements data anonymization.</p>

docs/modelarts/umn/modelarts_01_0013.html

@@ -1,10 +1,10 @@
 <a name="modelarts_01_0013"></a><a name="modelarts_01_0013"></a>
 
 <h1 class="topictitle1">DevEnviron</h1>
-<div id="body8662426"><p id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_p8060118">It is challenging to set up a development environment, select an AI algorithm framework and algorithm, debug code, install software, and accelerate hardware. To address these challenges, ModelArts provides DevEnviron to simplify the entire development process. </p>
+<div id="body8662426"><p id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_p8060118">It is challenging to set up a development environment, select an AI algorithm framework and algorithm, debug code, install software, and accelerate hardware. To address these challenges, ModelArts provides DevEnviron to simplify the entire development process.</p>
 <div class="fignone" id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_fig47081927202714"><span class="figcap"><b>Figure 1 </b>Algorithm development</span><br><span><img id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_image10386851143112" src="en-us_image_0000001110920904.png"></span></div>
 <ul id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_ul1590216418287"><li id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_li69020492816"><strong id="modelarts_01_0013__b19699172011719">Mainstream AI algorithm frameworks supported</strong><p id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_p21391717447">In the machine learning and deep learning fields, popular open-source training and inference frameworks include TensorFlow, PyTorch, MXNet, and MindSpore. ModelArts supports all popular AI computing frameworks and provides a user-friendly development and debugging environment. It supports traditional machine learning algorithms, such as logistic regression, decision tree, and clustering, as well as multiple types of deep learning algorithms, such as the convolutional neural network (CNN), recurrent neural network (RNN), and long short-term memory (LSTM).</p>
-</li><li id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_li55621235172819"><strong id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_b1759017361316">Simplified algorithm development for distributed training</strong><p id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_p15240133642813">Deep learning generally requires large-scale GPU clusters for distributed acceleration. For existing open-source frameworks, algorithm developers need to write a large amount of code for distributed training on different hardware, and the acceleration code varies depending on the framework. To resolve these issues, a distributed lightweight framework or SDK is required. The framework or SDK is built on deep learning engines such as TensorFlow, PyTorch, MXNet, and MindSpore to improve the distributed performance and usability of these engines. ModelArts MoXing perfectly suits the needs. The easy-to-use MoXing API/SDK enables you to develop deep learning at low costs. </p>
+</li><li id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_li55621235172819"><strong id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_b1759017361316">Simplified algorithm development for distributed training</strong><p id="modelarts_01_0013__en-us_topic_0284258739_en-us_topic_0168462754_p15240133642813">Deep learning generally requires large-scale GPU clusters for distributed acceleration. For existing open-source frameworks, algorithm developers need to write a large amount of code for distributed training on different hardware, and the acceleration code varies depending on the framework. To resolve these issues, a distributed lightweight framework or SDK is required. The framework or SDK is built on deep learning engines such as TensorFlow, PyTorch, MXNet, and MindSpore to improve the distributed performance and usability of these engines. ModelArts MoXing perfectly suits the needs. The easy-to-use MoXing API/SDK enables you to develop deep learning at low costs.</p>
 </li></ul>
 </div>
 <div>

docs/modelarts/umn/modelarts_01_0015.html

@@ -1,11 +1,10 @@
 <a name="modelarts_01_0015"></a><a name="modelarts_01_0015"></a>
 
 <h1 class="topictitle1">Model Deployment</h1>
-<div id="body8662426"><p id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_p1814018411037"></p>
+<div id="body8662426"><p id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_p16698024144411">Generally, AI model deployment and large-scale implementation are complex.</p>
-<p id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_p16698024144411">Generally, AI model deployment and large-scale implementation are complex.</p>
 <p id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_p1850322719445">ModelArts resolves this issue by deploying a trained model on different devices in various scenarios with only a few clicks. This secure and reliable one-stop deployment is available for individual developers, enterprises, and device manufacturers.</p>
 <div class="fignone" id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_fig104181892237"><span class="figcap"><b>Figure 1 </b>Process of deploying a model</span><br><span><img id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_image169391454237" src="en-us_image_0000001110920824.png"></span></div>
-<ul id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_ul814611421055"><li id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_li11461442653">The real-time inference service features high concurrency, low latency, and elastic scaling. </li><li id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_li86251938198">Models can be deployed as real-time inference services and batch inference tasks.</li></ul>
+<ul id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_ul814611421055"><li id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_li11461442653">The real-time inference service features high concurrency, low latency, and elastic scaling.</li><li id="modelarts_01_0015__en-us_topic_0284258741_en-us_topic_0168462756_li86251938198">Models can be deployed as real-time inference services and batch inference tasks.</li></ul>
 </div>
 <div>
 <div class="familylinks">

docs/modelarts/umn/modelarts_05_0003.html

@@ -1,11 +1,13 @@
 <a name="modelarts_05_0003"></a><a name="modelarts_05_0003"></a>
 
-<h1 class="topictitle1">What Are the Relationships Between ModelArts and Other Services</h1>
+<h1 class="topictitle1">What Are The Relationships Between ModelArts And Other Services</h1>
-<div id="body8662426"><div class="section" id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_section299615511534"><h4 class="sectiontitle">OBS</h4><p id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_p493914509919">ModelArts uses Object Storage Service (OBS) to store data and model backups and snapshots. OBS provides secure, reliable, low-cost storage. For more details, see <em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i3104928151115"></em><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i52511656161117"></em><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i1131122811129"><a href="https://docs.otc.t-systems.com/en-us/obs/index.html" target="_blank" rel="noopener noreferrer">Object Storage Service Console Function Overview</a></em><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i1919110719131"></em>.</p>
+<div id="body32001227"><div class="section" id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_section159941955105319"><h4 class="sectiontitle">IAM</h4><p id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_p11840154165516">ModelArts uses Identity and Access Management (IAM) for authentication and authorization. For more information about IAM, see <em id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_i569615374460"><a href="https://docs.otc.t-systems.com/en-us/iam/index.html" target="_blank" rel="noopener noreferrer">Identity and Access Management User Guide</a></em>.</p>
 </div>
-<div class="section" id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_section499755585318"><h4 class="sectiontitle">CCE</h4><p id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_p152905751010">ModelArts uses Cloud Container Engine (CCE) to deploy models as real-time services. CCE enables high concurrency and provides elastic scaling. For more information about CCE, see <u id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_u8895194525010"></u><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i225319514504"></em><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i179065818512"><a href="https://docs.otc.t-systems.com/en-us/cce/index.html" target="_blank" rel="noopener noreferrer">Cloud Container Engine User Guide</a></em>.</p>
+<div class="section" id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_section299615511534"><h4 class="sectiontitle">OBS</h4><p id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_p493914509919">ModelArts uses Object Storage Service (OBS) to store data and model backups and snapshots. OBS provides secure, reliable, low-cost storage. For more details, see <em id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_i1131122811129"><a href="https://docs.otc.t-systems.com/en-us/obs/index.html" target="_blank" rel="noopener noreferrer">Object Storage Service Console Function Overview</a></em><em id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_i1919110719131"></em>.</p>
 </div>
-<div class="section" id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_section872515283144"><h4 class="sectiontitle">SWR</h4><p id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_p518511154151">To use an AI framework that is not supported by ModelArts, use SoftWare Repository for Container (SWR) to customize an image and import the image to ModelArts for training or inference. For more details, see <em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i1715418119248"></em><em id="modelarts_05_0003__en-us_topic_0000001096750257_en-us_topic_0285164813_en-us_topic_0133336032_en-us_topic_0129439809_i11296534142412"></em>.</p>
+<div class="section" id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_section499755585318"><h4 class="sectiontitle">CCE</h4><p id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_p152905751010">ModelArts uses Cloud Container Engine (CCE) to deploy models as real-time services. CCE enables high concurrency and provides elastic scaling. For more information about CCE, see <em id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_i179065818512"><a href="https://docs.otc.t-systems.com/en-us/cce/index.html" target="_blank" rel="noopener noreferrer">Cloud Container Engine User Guide</a></em>.</p>
+</div>
+<div class="section" id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_section872515283144"><h4 class="sectiontitle">SWR</h4><p id="modelarts_05_0003__modelarts_01_0006_en-us_topic_0284258744_en-us_topic_0129439809_p518511154151">To use an AI framework that is not supported by ModelArts, use SoftWare Repository for Container (SWR) to customize an image and import the image to ModelArts for training or inference. For more details, see <em id="modelarts_05_0003__modelarts_01_0006_i11943153718013"><a href="https://docs.otc.t-systems.com/en-us/swr/index.html" target="_blank" rel="noopener noreferrer">SoftWare Repository for Container User Guide</a></em>.</p>
 </div>
 </div>
 <div>

docs/modelarts/umn/modelarts_05_0012.html

@@ -1,7 +1,8 @@
 <a name="modelarts_05_0012"></a><a name="modelarts_05_0012"></a>
 
 <h1 class="topictitle1">What Types of Services Can Models Be Deployed as on ModelArts?</h1>
-<div id="body8662426"></div>
+<div id="body8662426"><p id="modelarts_05_0012__en-us_topic_0000001096606441_en-us_topic_0285164862_en-us_topic_0139238484_p16231185042616">Currently, models can only be deployed as real-time services and batch services.</p>
+</div>
 <div>
 <div class="familylinks">
 <div class="parentlink"><strong>Parent topic:</strong> <a href="modelarts_05_0017.html">Service Deployment</a></div>

docs/modelarts/umn/modelarts_05_0013.html

@@ -1,7 +1,7 @@
 <a name="modelarts_05_0013"></a><a name="modelarts_05_0013"></a>
 
 <h1 class="topictitle1">How Do I Upload Data to OBS?</h1>
-<div id="body8662426"><p id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_p66197431013">Before using ModelArts to develop AI models, data needs to be uploaded to an OBS bucket. You can log in to the OBS console to create an OBS bucket, create a folder, and upload data. For details about how to upload data, see <em id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_i2576427123812"></em><em id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_i9969205518389"></em>.</p>
+<div id="body8662426"><p id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_p66197431013">Before using ModelArts to develop AI models, data needs to be uploaded to an OBS bucket. You can log in to the OBS console to create an OBS bucket, create a folder, and upload data. For details about how to upload data, see <em id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_i2576427123812"></em><em id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_i9969205518389"></em><em id="modelarts_05_0013__en-us_topic_0000001096750259_en-us_topic_0285164815_en-us_topic_0143299496_i69917743920"><a href="https://docs.otc.t-systems.com/usermanual/obs/obs_03_0403.html" target="_blank" rel="noopener noreferrer">Object Storage Service Getting Started</a></em>.</p>
 </div>
 <div>
 <div class="familylinks">

docs/modelarts/umn/modelarts_05_0014.html

@@ -6,7 +6,7 @@
 <ul class="ullinks">
 <li class="ulchildlink"><strong><a href="modelarts_05_0001.html">What Is ModelArts?</a></strong><br>
 </li>
-<li class="ulchildlink"><strong><a href="modelarts_05_0003.html">What Are the Relationships Between ModelArts and Other Services</a></strong><br>
+<li class="ulchildlink"><strong><a href="modelarts_05_0003.html">What Are The Relationships Between ModelArts And Other Services</a></strong><br>
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_05_0004.html">How Do I Obtain Access Keys?</a></strong><br>
 </li>
@@ -16,8 +16,6 @@
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0055.html">How Do I Use ModelArts to Train Models Based on Structured Data?</a></strong><br>
 </li>
-<li class="ulchildlink"><strong><a href="modelarts_21_0056.html">Why Cannot I Find the OBS Bucket on ModelArts After Uploading Data to OBS?</a></strong><br>
-</li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0057.html">Does ModelArts Support Multiple Projects?</a></strong><br>
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0058.html">How Do I View All Files in an OBS Directory on ModelArts?</a></strong><br>

docs/modelarts/umn/modelarts_05_0022.html

@@ -0,0 +1,30 @@
+<a name="modelarts_05_0022"></a><a name="modelarts_05_0022"></a>
+
+<h1 class="topictitle1">How Do I Install External Libraries in a Notebook Instance?</h1>
+<div id="body0000001305566093"><p id="modelarts_05_0022__en-us_topic_0165695111_p1439612494411">Multiple environments have been integrated into ModelArts Notebook. These environments contain Jupyter Notebook and Python packages, including TensorFlow, MXNet, Caffe, PyTorch, and Spark. You can use <strong id="modelarts_05_0022__en-us_topic_0165695111_b268637862">pip install</strong> to install external libraries in Jupyter Notebook or on the <strong id="modelarts_05_0022__en-us_topic_0165695111_b2955136115414">Terminal</strong> page.</p>
+<div class="section" id="modelarts_05_0022__en-us_topic_0165695111_section2244549752"><h4 class="sectiontitle">Installing External Libraries in Jupyter Notebook</h4><p id="modelarts_05_0022__en-us_topic_0165695111_p657548768">For example, use Jupyter Notebook to install Shapely in the <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname3899205613195"><b>TensorFlow-1.8</b></span> environment.</p>
+<ol id="modelarts_05_0022__en-us_topic_0165695111_ol1372110498286"><li id="modelarts_05_0022__en-us_topic_0165695111_li15722164915281">Open a notebook instance.</li><li id="modelarts_05_0022__en-us_topic_0165695111_li167221649142818">On the Jupyter Notebook dashboard, choose <strong id="modelarts_05_0022__en-us_topic_0165695111_b1787265435413">New</strong> &gt; <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname12980151213201"><b>TensorFlow-1.8</b></span>.</li><li id="modelarts_05_0022__en-us_topic_0165695111_li572254952814">In the new notebook instance, enter the following command in the code input bar:<p id="modelarts_05_0022__en-us_topic_0165695111_p8438163612203"><a name="modelarts_05_0022__en-us_topic_0165695111_li572254952814"></a><a name="en-us_topic_0165695111_li572254952814"></a><b><span class="cmdname" id="modelarts_05_0022__en-us_topic_0165695111_cmdname1122810396209">!pip install Shapely</span></b></p>
+</li></ol>
+</div>
+<div class="section" id="modelarts_05_0022__en-us_topic_0165695111_section53701401087"><h4 class="sectiontitle">Installing External Libraries on the <strong id="modelarts_05_0022__en-us_topic_0165695111_b20875229175614">Terminal</strong> Page</h4><p id="modelarts_05_0022__en-us_topic_0165695111_p910314415284">For example, use <b><span class="cmdname" id="modelarts_05_0022__en-us_topic_0165695111_cmdname59804911201">pip</span></b> to install Shapely in the <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname6544114562020"><b>TensorFlow-1.8</b></span> environment on the <strong id="modelarts_05_0022__en-us_topic_0165695111_b16624811584">Terminal</strong> page.</p>
+<ol id="modelarts_05_0022__en-us_topic_0165695111_ol54111544182813"><li id="modelarts_05_0022__en-us_topic_0165695111_li12411194442811">Open a notebook instance.</li><li id="modelarts_05_0022__en-us_topic_0165695111_li2041116442288">On the Jupyter Notebook dashboard, choose <strong id="modelarts_05_0022__en-us_topic_0165695111_b174259157582">New</strong> &gt; <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname145491312221"><b>Terminal</b></span>.</li><li id="modelarts_05_0022__en-us_topic_0165695111_li34121446284">Enter the following command in the code input bar to obtain the command for activating <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname20996161895919"><b>TensorFlow-1.8</b></span> and activate it:<p id="modelarts_05_0022__en-us_topic_0165695111_p48663382211"><b><span class="cmdname" id="modelarts_05_0022__en-us_topic_0165695111_cmdname1046857152212">cat /home/ma-user/README</span></b></p>
+<p id="modelarts_05_0022__en-us_topic_0165695111_p886113316222"><b><span class="cmdname" id="modelarts_05_0022__en-us_topic_0165695111_cmdname1968135752213">source /home/ma-user/anaconda3/bin/activate TensorFlow-1.8</span></b></p>
+<div class="note" id="modelarts_05_0022__en-us_topic_0165695111_note19856951102518"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><p id="modelarts_05_0022__en-us_topic_0165695111_p9857651172512">If you use another engine, replace <span class="parmname" id="modelarts_05_0022__en-us_topic_0165695111_parmname1292123012912"><b>TensorFlow-1.8</b></span> in the command with the name and version of the engine.</p>
+</div></div>
+<div class="fignone" id="modelarts_05_0022__en-us_topic_0165695111_fig19791052144319"><span class="figcap"><b>Figure 1 </b>Activating the environment</span><br><span><img id="modelarts_05_0022__en-us_topic_0165695111_image1880145220435" src="en-us_image_0000001281686748.png"></span></div>
+</li><li id="modelarts_05_0022__en-us_topic_0165695111_li1412124413287">Type the following command in the code input bar to install Shapely:<p id="modelarts_05_0022__en-us_topic_0165695111_p14886133620467"><a name="modelarts_05_0022__en-us_topic_0165695111_li1412124413287"></a><a name="en-us_topic_0165695111_li1412124413287"></a><b><span class="cmdname" id="modelarts_05_0022__en-us_topic_0165695111_cmdname587985817230">pip install Shapely</span></b></p>
+</li></ol>
+<div class="note" id="modelarts_05_0022__en-us_topic_0165695111_note4186132135512"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><p id="modelarts_05_0022__en-us_topic_0165695111_p278538105520">A new independent running environment is opened when a ModelArts training job is created. The new environment is not associated with the packages installed in the notebook environment. Therefore, add <strong id="modelarts_05_0022__en-us_topic_0165695111_b1753915418219">os.system('pip install <em id="modelarts_05_0022__i6527942210">xxx</em>')</strong> to the boot code before importing the installation package.</p>
+<p id="modelarts_05_0022__en-us_topic_0165695111_p197820386559">For example, if you need to use the Shapely dependency package in the training job, add the following code to the boot code after the notebook instance is installed:</p>
+<pre class="screen" id="modelarts_05_0022__screen7701511152415">import os
+os.system('pip install Shapely')
+import Shapely</pre>
+</div></div>
+</div>
+</div>
+<div>
+<div class="familylinks">
+<div class="parentlink"><strong>Parent topic:</strong> <a href="modelarts_05_0067.html">Notebook</a></div>
+</div>
+</div>
+

docs/modelarts/umn/modelarts_05_0024.html

@@ -0,0 +1,56 @@
+<a name="modelarts_05_0024"></a><a name="modelarts_05_0024"></a>
+
+<h1 class="topictitle1">How Do I Upload a File from a Notebook Instance to OBS or Download a File from OBS to a Notebook Instance?</h1>
+<div id="body32001227"><p id="modelarts_05_0024__en-us_topic_0165695113_p174675641814">In a notebook instance, you can call the ModelArts MoXing API or SDK to exchange data with OBS for uploading a file to OBS or downloading a file from OBS to the notebook instance.</p>
+<div class="fignone" id="modelarts_05_0024__en-us_topic_0165695113_fig15668759124612"><span class="figcap"><b>Figure 1 </b>Uploading or downloading a file</span><br><span><img id="modelarts_05_0024__en-us_topic_0165695113_image1420817011019" src="en-us_image_0000001290603082.png"></span></div>
+<div class="section" id="modelarts_05_0024__en-us_topic_0165695113_section1470917513359"><h4 class="sectiontitle">Method 1: Using MoXing to upload and download a file</h4><p id="modelarts_05_0024__en-us_topic_0165695113_p1614312534213">Developed by the ModelArts team, MoXing is a distributed training acceleration framework built on open-source deep learning engines such as TensorFlow and PyTorch. MoXing makes model coding easier and more efficient.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p15143165318214">MoXing provides a set of file object APIs for reading and writing OBS files.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p7177185611211">Sample code:</p>
+<pre class="screen" id="modelarts_05_0024__en-us_topic_0165695113_screen69336565356">import moxing as mox
+
+# Download the OBS folder <strong id="modelarts_05_0024__en-us_topic_0165695113_b1973345963717">sub_dir_0</strong> from OBS to a notebook instance.
+mox.file.copy_parallel('obs://bucket_name/sub_dir_0', '/home/ma-user/work/sub_dir_0')
+# Download the OBS file <strong id="modelarts_05_0024__en-us_topic_0165695113_b19476142389">obs_file.txt</strong> from OBS to a notebook instance.
+mox.file.copy('obs://bucket_name/obs_file.txt', '/home/ma-user/work/obs_file.txt')
+
+# Upload the OBS folder <strong id="modelarts_05_0024__en-us_topic_0165695113_b321810410381">sub_dir_0</strong> from a notebook instance to OBS.
+mox.file.copy_parallel('/home/ma-user/work/sub_dir_0', 'obs://bucket_name/sub_dir_0')
+# Upload the OBS file <strong id="modelarts_05_0024__en-us_topic_0165695113_b17820181363912">obs_file.txt</strong> from a notebook instance to OBS.
+mox.file.copy('/home/ma-user/work/obs_file.txt', 'obs://bucket_name/obs_file.txt')</pre>
+</div>
+<div class="section" id="modelarts_05_0024__en-us_topic_0165695113_section16382288170"><h4 class="sectiontitle">Method 2: Using SDK to upload and download a file</h4><p id="modelarts_05_0024__en-us_topic_0165695113_p185911632141811">Call the ModelArts SDK for downloading a file from OBS.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p10189113013186">Sample code: Download <strong id="modelarts_05_0024__en-us_topic_0165695113_b865945173418">file1.txt</strong> from OBS to <strong id="modelarts_05_0024__en-us_topic_0165695113_b1069011914393">/home/ma-user/work/</strong> in the notebook instance. All the bucket name, folder name, and file name are customizable.</p>
+<div class="codecoloring" codetype="Python" id="modelarts_05_0024__en-us_topic_0165695113_screen81894303188"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre>1
+2
+3</pre></div></td><td class="code"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">modelarts.session</span> <span class="kn">import</span> <span class="n">Session</span>
+<span class="n">session</span> <span class="o">=</span> <span class="n">Session</span><span class="p">()</span>
+<span class="n">session</span><span class="o">.</span><span class="n">obs</span><span class="o">.</span><span class="n">download_file</span><span class="p">(</span><span class="n">src_obs_file</span><span class="o">=</span><span class="s2">&quot;obs://bucket-name/dir1/file1.txt&quot;</span><span class="p">,</span> <span class="n">dst_local_dir</span><span class="o">=</span><span class="s2">&quot;/home/ma-user/work/&quot;</span><span class="p">)</span>
+</pre></div>
+</td></tr></table></div>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p1328616715524">Call the ModelArts SDK for downloading a folder from OBS.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p5899105445211">Sample code: Download <strong id="modelarts_05_0024__en-us_topic_0165695113_b1801420111016">dir1</strong> from OBS to <strong id="modelarts_05_0024__en-us_topic_0165695113_b10032071014">/home/ma-user/work/</strong> in the notebook instance. The bucket name and folder name are customizable.</p>
+<pre class="screen" id="modelarts_05_0024__en-us_topic_0165695113_screen152791249105319">from modelarts.session import Session
+session = Session()
+session.obs.download_dir(src_obs_dir="obs://bucket-name/dir1/", dst_local_dir="/home/ma-user/work/")</pre>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p111932714467">Call the ModelArts SDK for uploading a file to OBS.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p15879145745018">Sample code: Upload <strong id="modelarts_05_0024__en-us_topic_0165695113_b46291338101210">file1.txt</strong> in the notebook instance to OBS bucket <strong id="modelarts_05_0024__en-us_topic_0165695113_b2593630181320">obs://bucket-name/dir1/</strong>. All the bucket name, folder name, and file name are customizable.</p>
+<div class="codecoloring" codetype="Python" id="modelarts_05_0024__en-us_topic_0165695113_screen32018257496"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre>1
+2
+3</pre></div></td><td class="code"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">modelarts.session</span> <span class="kn">import</span> <span class="n">Session</span>
+<span class="n">session</span> <span class="o">=</span> <span class="n">Session</span><span class="p">()</span>
+<span class="n">session</span><span class="o">.</span><span class="n">obs</span><span class="o">.</span><span class="n">upload_file</span><span class="p">(</span><span class="n">src_local_file</span><span class="o">=</span><span class="s1">'/home/ma-user/work/file1.txt'</span><span class="p">,</span> <span class="n">dst_obs_dir</span><span class="o">=</span><span class="s1">'obs://bucket-name/dir1/'</span><span class="p">)</span>
+</pre></div>
+</td></tr></table></div>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p181118260554">Call the ModelArts SDK for uploading a folder to OBS.</p>
+<p id="modelarts_05_0024__en-us_topic_0165695113_p1781112611552">Sample code: Upload <strong id="modelarts_05_0024__en-us_topic_0165695113_b763113451517">/work/</strong> in the notebook instance to <strong id="modelarts_05_0024__en-us_topic_0165695113_b749155271620">obs://bucket-name/dir1/work/</strong> of <strong id="modelarts_05_0024__en-us_topic_0165695113_b5445214141612">bucket-name</strong>. The bucket name and folder name are customizable.</p>
+<pre class="screen" id="modelarts_05_0024__en-us_topic_0165695113_screen13811202613553">from modelarts.session import Session
+session = Session()
+session.obs.upload_dir(src_local_dir='/home/ma-user/work/', dst_obs_dir='obs://bucket-name/dir1/')</pre>
+</div>
+</div>
+<div>
+<div class="familylinks">
+<div class="parentlink"><strong>Parent topic:</strong> <a href="modelarts_05_0067.html">Notebook</a></div>
+</div>
+</div>
+

docs/modelarts/umn/modelarts_05_0032.html

@@ -6,7 +6,7 @@
 <div class="section" id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_section459111152018"><h4 class="sectiontitle">Solution</h4><ol id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_ol1685031692911"><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li785091618292">Confirm that the OBS path in the log exists.<p id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_p13646204181114"><a name="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li785091618292"></a><a name="en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li785091618292"></a>Locate the incorrect OBS path in the log, for example, <span class="filepath" id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_filepath192420541552"><b>obs-test/ModelArts/examples/</b></span>. There are two methods to check whether it exists.</p>
 <ul id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_ul1998720101215"><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li898112015127">On OBS Console, check whether the OBS path exists.<p id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_p15591145510124"><a name="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li898112015127"></a><a name="en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li898112015127"></a>Log in to OBS console using the current account, and check whether the OBS buckets, folders, and files exist in the OBS path displayed in the log. For example, you can confirm that a given bucket is there and then check if that bucket contains the folder you are looking for based on the configured path.</p>
 <ul id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_ul2594812141517"><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li1539941114157">If the file path exists, go to <a href="#modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li77081222112915">2</a>.</li><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li1419101612157">If it does not exist, change the path configured for the training job to an OBS bucket path that is actually there.</li></ul>
-</li><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li18436154511129">Create a notebook instance, and use an API to check whether the directory exists. In an existing notebook instance or after creating a new notebook instance, run the following command to check whether the directory exists:<pre class="screen" id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_screen667275481615">import moxing as mox
+</li><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li18436154511129">Create a notebook instance, and use an API to check whether the directory exists. In an existing notebook instance or after creating a new notebook instance, run the following command to check whether the directory exists:<pre class="screen" id="modelarts_05_0032__screen316416232254">import moxing as mox
 mox.file.exists('obs://obs-test/ModelArts/examples/')</pre>
 <ul id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_ul10694527201713"><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li16694192718173">If it exists, go to <a href="#modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li77081222112915">2</a>.</li><li id="modelarts_05_0032__en-us_topic_0000001096606439_en-us_topic_0285164857_en-us_topic_0166743701_li156947272172">If it does not exist, change it to an available OBS bucket path in the training job.</li></ul>
 </li></ul>

docs/modelarts/umn/modelarts_05_0063.html

@@ -6,21 +6,21 @@
 </div></div>
 <p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p1051702281913">Create a file named <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath1551719224198"><b>pip-requirements.txt</b></span> in the code directory. In this file, specify the name and version of the dependency package in the format of <em id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_i094863161910">Package name</em><strong id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_b1594818311199">==</strong><em id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_i18948034197">Version</em>.</p>
 <p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p14517922181914">For example, the OBS path specified by <span class="parmname" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_parmname12230141822012"><b>Code Directory</b></span> contains model files and the <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath18517192220193"><b>pip-requirements.txt</b></span> file. The following shows the code directory structure:</p>
-<pre class="screen" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_screen951718224198">|---OBS path to the model boot file
+<pre class="screen" id="modelarts_05_0063__screen0019422258">|---OBS path to the model boot file
      |---model.py               #Model boot file
      |---pip-requirements.txt   #Customized configuration file, which specifies the name and version of the dependency package</pre>
 <p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p3517132221911">The following shows the content of the <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath6517122151919"><b>pip-requirements.txt</b></span> file:</p>
-<pre class="screen" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_screen7517112211917">alembic==0.8.6
+<pre class="screen" id="modelarts_05_0063__screen8414185592518">alembic==0.8.6
 bleach==1.4.3
 click==6.6</pre>
 </li><li id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_li106572405813"><strong id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_b1782510249238">Customized WHL file</strong><p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p72677538812">When you use a customized .whl file, the system cannot automatically download and install the file. Place the .whl file in the code directory, create a file named <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath82081126592"><b>pip-requirements.txt</b></span>, and specify the name of the .whl file in the created file. The dependency package must be a .whl file.</p>
 <p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p8517195916108">For example, the OBS path specified by <span class="parmname" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_parmname2391541112517"><b>Code Directory</b></span> contains model files, .whl file, and <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath05161459191012"><b>pip-requirements.txt</b></span> file. The following shows the code directory structure:</p>
-<pre class="screen" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_screen0831354101214">|---OBS path to the model boot file
+<pre class="screen" id="modelarts_05_0063__screen157475104267">|---OBS path to the model boot file
      |---model.py               #Model boot file
      |---XXX.whl                #Dependency package. If multiple dependencies are required, place all of them here.
      |---pip-requirements.txt   #Customized configuration file, which specifies the name of the dependency package</pre>
 <p id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_p921092518913">The following shows the content of the <span class="filepath" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_filepath86762393142"><b>pip-requirements.txt</b></span> file:</p>
-<pre class="screen" id="modelarts_05_0063__en-us_topic_0000001096467413_en-us_topic_0285164858_en-us_topic_0186387866_screen1749471118109">numpy-1.15.4-cp36-cp36m-manylinux1_x86_64.whl
+<pre class="screen" id="modelarts_05_0063__screen12399162520260">numpy-1.15.4-cp36-cp36m-manylinux1_x86_64.whl
 tensorflow-1.8.0-cp36-cp36m-manylinux1_x86_64.whl</pre>
 </li></ul>
 </div>

docs/modelarts/umn/modelarts_05_0067.html

@@ -6,7 +6,7 @@
 <ul class="ullinks">
 <li class="ulchildlink"><strong><a href="modelarts_05_0071.html">How Do I Enable the Terminal Function in DevEnviron of ModelArts?</a></strong><br>
 </li>
-<li class="ulchildlink"><strong><a href="modelarts_21_0064.html">How Do I Install External Libraries in a Notebook Instance?</a></strong><br>
+<li class="ulchildlink"><strong><a href="modelarts_05_0022.html">How Do I Install External Libraries in a Notebook Instance?</a></strong><br>
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0065.html">Is the Keras Engine Supported?</a></strong><br>
 </li>
@@ -14,6 +14,8 @@
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0067.html">What Should I Do When the System Displays an Error Message Indicating that No Space Left After I Run the pip install Command?</a></strong><br>
 </li>
+<li class="ulchildlink"><strong><a href="modelarts_05_0024.html">How Do I Upload a File from a Notebook Instance to OBS or Download a File from OBS to a Notebook Instance?</a></strong><br>
+</li>
 <li class="ulchildlink"><strong><a href="modelarts_21_0068.html">How Do I Upload Local Files to a Notebook Instance?</a></strong><br>
 </li>
 <li class="ulchildlink"><strong><a href="modelarts_05_0045.html">Where Will the Data Be Uploaded to?</a></strong><br>

docs/modelarts/umn/modelarts_05_0081.html

@@ -1,7 +1,7 @@
 <a name="modelarts_05_0081"></a><a name="modelarts_05_0081"></a>
 
 <h1 class="topictitle1">Where Is Data Stored After the Sync OBS Function Is Used?</h1>
-<div id="body8662426"><ol id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_ol1916513255134"><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li6683105013126">Log in to the ModelArts management console, and choose <span class="menucascade" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_menucascade122641384146"><b><span class="uicontrol" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_uicontrol11262143818148">DevEnviron &gt; Notebooks</span></b></span>.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li1521334591415">In the <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname61412722418"><b>Operation</b></span> column of the target notebook instance in the notebook list, click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname5142474242"><b>Open</b></span> to go to the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle1914257202412"><b>Jupyter</b></span> page.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li107718439535">On the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle197914375319"><b>Files</b></span> tab page of the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle7961953202610"><b>Jupyter</b></span> page, select the target file and click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname479943205319"><b>Sync OBS</b></span> in the upper part of the page to synchronize the file. The file is stored in the <span class="filepath" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_filepath154696793816"><b>~/work</b></span> directory of the instance.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li5113201812176">On the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle18599832192713"><b>Files</b></span> tab page of the <strong id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_b99141436174118">Jupyter</strong> page, click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname5601133215275"><b>New</b></span> and select <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname15603332102713"><b>Terminal</b></span>. The <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname96047325271"><b>Terminal</b></span> page is displayed.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li13212849173814">Run the following command to go to the <span class="filepath" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_filepath2832133814409"><b>~/work</b></span> directory.<pre class="screen" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_screen151831492382">cd work</pre>
+<div id="body8662426"><ol id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_ol1916513255134"><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li6683105013126">Log in to the ModelArts management console, and choose <span class="menucascade" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_menucascade122641384146"><b><span class="uicontrol" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_uicontrol11262143818148">DevEnviron &gt; Notebooks</span></b></span>.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li1521334591415">In the <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname61412722418"><b>Operation</b></span> column of the target notebook instance in the notebook list, click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname5142474242"><b>Open</b></span> to go to the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle1914257202412"><b>Jupyter</b></span> page.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li107718439535">On the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle197914375319"><b>Files</b></span> tab page of the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle7961953202610"><b>Jupyter</b></span> page, select the target file and click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname479943205319"><b>Sync OBS</b></span> in the upper part of the page to synchronize the file. The file is stored in the <span class="filepath" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_filepath154696793816"><b>~/work</b></span> directory of the instance.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li5113201812176">On the <span class="wintitle" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_wintitle18599832192713"><b>Files</b></span> tab page of the <strong id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_b99141436174118">Jupyter</strong> page, click <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname5601133215275"><b>New</b></span> and select <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname15603332102713"><b>Terminal</b></span>. The <span class="parmname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_parmname96047325271"><b>Terminal</b></span> page is displayed.</li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li13212849173814">Run the following command to go to the <span class="filepath" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_filepath2832133814409"><b>~/work</b></span> directory.<pre class="screen" id="modelarts_05_0081__screen15226105132413">cd work</pre>
 </li><li id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_li511161818171">Run the <b><span class="cmdname" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_cmdname11362625111811">ls</span></b> command in the <span class="filepath" id="modelarts_05_0081__en-us_topic_0000001096750265_en-us_topic_0285164854_en-us_topic_0198037462_filepath18906344174020"><b>~/work</b></span> directory to view the files.</li></ol>
 </div>
 <div>

docs/modelarts/umn/modelarts_05_0090.html

@@ -1,7 +1,7 @@
 <a name="modelarts_05_0090"></a><a name="modelarts_05_0090"></a>
 
 <h1 class="topictitle1">What Are Sizes of the /cache Directories for Different Resource Specifications in the Training Environment?</h1>
-<div id="body8662426"><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p52514553619">When creating a training job, you can select CPU, GPU, or Ascend resources based on the size of the training job.</p>
+<div id="body8662426"><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p52514553619">When creating a training job, you can select CPU, GPU<span id="modelarts_05_0090__ph1243555591512"> </span>resources based on the size of the training job.</p>
 <p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p1352593019279">ModelArts mounts the disk to the <strong id="modelarts_05_0090__en-us_topic_0000001096855431_b919491193714">/cache</strong> directory. You can use this directory to store temporary files. The <strong id="modelarts_05_0090__en-us_topic_0000001096855431_b98621238372">/cache</strong> directory shares resources with the code directory. The directory has different capacities for different resource specifications.</p>
 <ul id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_ul71421635862"><li id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_li814218351561">GPU resources
 <div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_table9533182215915" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Capacities of the cache directories for GPU resources</caption><thead align="left"><tr id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_row7534022115910"><th align="left" class="cellrowborder" valign="top" width="22.28%" id="mcps1.3.3.1.1.2.3.1.1"><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p353482265914">GPU Specifications</p>
@@ -49,22 +49,6 @@
 </table>
 </div>
 </li></ul>
-<ul id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_ul16388103291919"><li id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_li14388173212195">Ascend resources
-<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_table1652218538206" frame="border" border="1" rules="all"><caption><b>Table 3 </b>Capacities of the cache directories for Ascend resources</caption><thead align="left"><tr id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_row1652265382010"><th align="left" class="cellrowborder" valign="top" width="25.09%" id="mcps1.3.4.1.1.2.3.1.1"><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p19522253112017">Ascend Specifications</p>
-</th>
-<th align="left" class="cellrowborder" valign="top" width="74.91%" id="mcps1.3.4.1.1.2.3.1.2"><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p35228539204">cache Directory Capacity</p>
-</th>
-</tr>
-</thead>
-<tbody><tr id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_row115221153162012"><td class="cellrowborder" valign="top" width="25.09%" headers="mcps1.3.4.1.1.2.3.1.1 "><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p152205316203">Ascend 910</p>
-</td>
-<td class="cellrowborder" valign="top" width="74.91%" headers="mcps1.3.4.1.1.2.3.1.2 "><p id="modelarts_05_0090__en-us_topic_0000001096855431_en-us_topic_0200343601_p852385313201">3T</p>
-</td>
-</tr>
-</tbody>
-</table>
-</div>
-</li></ul>
 </div>
 <div>
 <div class="familylinks">

docs/modelarts/umn/modelarts_05_0128.html

@@ -114,7 +114,7 @@
 </th>
 <th align="left" class="cellrowborder" valign="top" width="23.232323232323235%" id="mcps1.3.3.3.2.7.1.5"><p id="modelarts_05_0128__en-us_topic_0216621183_en-us_topic_0298539662_p179127131467">AI Engine and Version</p>
 </th>
-<th align="left" class="cellrowborder" valign="top" width="18.181818181818183%" id="mcps1.3.3.3.2.7.1.6"><p id="modelarts_05_0128__en-us_topic_0216621183_en-us_topic_0298539662_p48709364133">Supported CUDA or Ascend Version</p>
+<th align="left" class="cellrowborder" valign="top" width="18.181818181818183%" id="mcps1.3.3.3.2.7.1.6"><p id="modelarts_05_0128__en-us_topic_0216621183_en-us_topic_0298539662_p48709364133">Supported CUDA<span id="modelarts_05_0128__ph374542419134"> </span> Version</p>
 </th>
 </tr>
 </thead>

docs/modelarts/umn/modelarts_08_0002.html

@@ -5,6 +5,7 @@
 <div class="section" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_section13798121135718"><h4 class="sectiontitle">Obtaining an Access Key</h4><ol id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_ol18857141211570"><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li1085711128572">On the ModelArts management console, hover over the username in the upper right corner and choose <span class="parmname" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_parmname12881354155712"><b>My Credentials</b></span> from the drop-down list.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li737152217588">On the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b787613444318">My Credentials</strong> page, choose <span class="menucascade" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_menucascade118931943436"><b><span class="uicontrol" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_uicontrol789113418435">Access Keys</span></b> &gt; <b><span class="uicontrol" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_uicontrol1589244164318">Create Access Key</span></b></span>.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li33852235815">In the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b184271189432">Create Access Key</strong> dialog box that is displayed, enter the verification code received by SMS or email.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li1738322115819">Click <span class="uicontrol" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_uicontrol195220462582"><b>OK</b></span> and save the access key file as prompted. The access key file is saved in the default download folder of the browser. Open the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b1026996972">credentials.csv</strong> file to view the access key (<strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b2044922722">Access Key Id</strong> and <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b1285491740">Secret Access Key</strong>).</li></ol>
 </div>
 <div class="section" id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_section9701429311"><h4 class="sectiontitle">Adding an Access Key</h4><ol id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_ol194759579116"><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li54753571218">Log in to the ModelArts management console. In the navigation pane, choose <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b206121430152117">Settings</strong>. The <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b261353012117">Settings</strong> page is displayed.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li23519440165">Click <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b1611671505518">Add Authorization</strong>.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li1583984172">In the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b192769515582">Add Authorization</strong> dialog box that is displayed, set <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b0428859175812">Authorization Method</strong> to <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b8789171316599">AK/SK</strong>. The username is fixed. Enter the obtained access key pair.<ul id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_ul14551873217"><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li55518717210"><strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b058594424617">AK</strong>: Enter the value of the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b9587174424611">Access Key Id</strong> field in the key file.</li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li45511078216"><strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b16519114617460">SK</strong>: Enter the value of the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b11520104611466">Secret Access Key</strong> field in the key file.</li></ul>
+<p id="modelarts_08_0002__p147201948131111"><span><img id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_image8978824185314" src="en-us_image_0000001156920825.png"></span></p>
 </li><li id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_li5551471525">Select <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b1719661125612">I have read and agree to the ModelArts Service Statement</strong> and click <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b12355713195614">Agree</strong>.<p id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_p1010863219487">After the configuration is complete, you can view the access key configurations of an account or IAM user on the <strong id="modelarts_08_0002__en-us_topic_0284258828_en-us_topic_0256240292_b1272474035610">Settings</strong> page.</p>
 </li></ol>
 </div>

docs/modelarts/umn/modelarts_08_0003.html

@@ -2,9 +2,9 @@
 
 <h1 class="topictitle1">Creating an OBS Bucket</h1>
 <div id="body8662426"><p id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_p12169122719476">ModelArts uses OBS to store data and model backups and snapshots, achieving secure, reliable, and low-cost storage. Therefore, before using ModelArts, create an OBS bucket and folders for storing data.</p>
-<div class="section" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_section763817416543"><h4 class="sectiontitle">Procedure</h4><ol id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_ol158321616185816"><li id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_li11270172211210">Log in to OBS Console and create an OBS bucket. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/obs/en-us_topic_0045853662.html" target="_blank" rel="noopener noreferrer">Creating a Bucket</a>. For example, create an OBS bucket named <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath1422122115019"><b>c-flowers</b></span>.<div class="note" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_note182501262482"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><p id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_p19251126144814">The created OBS bucket and ModelArts are in the same region. </p>
+<div class="section" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_section763817416543"><h4 class="sectiontitle">Procedure</h4><ol id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_ol158321616185816"><li id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_li11270172211210">Log in to OBS Console and create an OBS bucket. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/obs/en-us_topic_0045853662.html" target="_blank" rel="noopener noreferrer">Creating a Bucket</a>. For example, create an OBS bucket named <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath1422122115019"><b>c-flowers</b></span>.<div class="note" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_note182501262482"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><p id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_p19251126144814">The created OBS bucket and ModelArts are in the same region.Create a folder for storing data. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/obs/obs_03_0316.html" target="_blank" rel="noopener noreferrer">Creating a Folder</a>. For example, create a folder named <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath14774153018017"><b>flowers</b></span> in the created <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath103706017018"><b>c-flowers</b></span> OBS bucket.</p>
 </div></div>
-</li><li id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_li191305718156">Create a folder for storing data. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/obs/obs_03_0316.html" target="_blank" rel="noopener noreferrer">Creating a Folder</a>. For example, create a folder named <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath14774153018017"><b>flowers</b></span> in the created <span class="filepath" id="modelarts_08_0003__en-us_topic_0284258829_en-us_topic_0168522442_filepath103706017018"><b>c-flowers</b></span> OBS bucket.</li></ol>
+</li></ol>
 </div>
 </div>
 <div>

docs/modelarts/umn/modelarts_08_0005.html

@@ -3,7 +3,7 @@
 <h1 class="topictitle1">Overview</h1>
 <div id="body8662426"><p id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_p15539163118913">When you use ExeML, data management, notebook instances, training jobs, models, and services, ModelArts may need to access dependent services such as OBS and Software Repository for Container (SWR). If ModelArts is not authorized to access the services, these functions cannot be used.</p>
 <p id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_p6338138201017">You can configure access authorization in either of the following ways:</p>
-<ul id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_ul2311182521017"><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li3311122591014"><strong id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_b4705526760">Using an agency</strong> (recommended)<p id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_p8272412103">After agency authorization is configured, the dependent service operation permissions are delegated to ModelArts so that ModelArts can use the dependent services and perform operations on resources on your behalf. </p>
+<ul id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_ul2311182521017"><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li3311122591014"><strong id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_b4705526760">Using an agency</strong> (recommended)<p id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_p8272412103">After agency authorization is configured, the dependent service operation permissions are delegated to ModelArts so that ModelArts can use the dependent services and perform operations on resources on your behalf.</p>
 </li><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li01711354102"><strong id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_b1211919595532">Using the access key</strong><p id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_p1938195651012">You can use the obtained access key pair (AK/SK) to authorize ModelArts to access dependent services and and perform operations on resources.</p>
 </li></ul>
 <div class="section" id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_section380972635315"><h4 class="sectiontitle">Precautions</h4><ul id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_ul19637173345314"><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li196370339538">Agency authorization grants ModelArts permissions on dependent services, such as OBS and SWR. If the OBS permissions are not configured for an IAM user, the user still does not have the permission to operate the services.</li><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li1987205315010">For users who have used ModelArts before, access key authorization has been configured and does not need to be configured again. However, you are advised to use agency authorization again.</li><li id="modelarts_08_0005__en-us_topic_0284258826_en-us_topic_0256240290_li2587135920433">For new users, use agency authorization.</li></ul>

docs/modelarts/umn/modelarts_08_0007.html

@@ -27,16 +27,16 @@
 </tr>
 <tr id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_row13954142161520"><td class="cellrowborder" valign="top" width="19.98%" headers="mcps1.3.4.2.3.4.2.3.1.1 "><p id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_p395442121515">Agency</p>
 </td>
-<td class="cellrowborder" valign="top" width="80.02%" headers="mcps1.3.4.2.3.4.2.3.1.2 "><ul id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_ul1850113131616"><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li850123201615"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b19894194919456">Auto Create</strong> (recommended): When you use ModelArts <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b6895049114519">Settings</strong> for the first time, no agency is available. In this case, you can click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b5895144912455">Auto Create</strong> to automatically create an agency for the user selected in <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b78951949124519">Username</strong>. For details about the automatically created agency, see <a href="#modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519">Automatically Creating an IAM Agency</a>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li105015321617">Select an existing agency: If you have created agencies in IAM, you can select an available agency from the drop-down list to authorize the selected user.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li3501163121614"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b17853112732019">Create on IAM</strong>: If the automatically created agency cannot meet your requirements, you can click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1530173832016">Create on IAM</strong> to manually create an agency on the IAM management console.  If you choose <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b5262161703315">Create on IAM</strong>, configure at least the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b17943162518336">ModelArts CommonOperation</strong> and <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b15482193033318">OBS Operate Access</strong> permissions. Otherwise, the basic functions of ModelArts will be unavailable.</li></ul>
+<td class="cellrowborder" valign="top" width="80.02%" headers="mcps1.3.4.2.3.4.2.3.1.2 "><ul id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_ul1850113131616"><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li850123201615"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b19894194919456">Auto Create</strong> (recommended): When you use ModelArts <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b6895049114519">Settings</strong> for the first time, no agency is available. In this case, you can click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b5895144912455">Auto Create</strong> to automatically create an agency for the user selected in <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b78951949124519">Username</strong>. For details about the automatically created agency, see <a href="#modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519">Automatically Creating an IAM Agency</a>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li105015321617">Select an existing agency: If you have created agencies in IAM, you can select an available agency from the drop-down list to authorize the selected user.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li3501163121614"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b17853112732019">Create on IAM</strong>: If the automatically created agency cannot meet your requirements, you can click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1530173832016">Create on IAM</strong> to manually create an agency on the IAM management console. If you choose <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b5262161703315">Create on IAM</strong>, configure at least the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b17943162518336">ModelArts CommonOperation</strong> and <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b15482193033318">OBS Operate Access</strong> permissions. Otherwise, the basic functions of ModelArts will be unavailable.</li></ul>
 </td>
 </tr>
 </tbody>
 </table>
 </div>
-</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li16660165611718">After configuring the username and agency, select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b04136444251">I have read and agree to the ModelArts Service Statement</strong> and click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b198971428172617">Agree</strong>.<p id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_p1010863219487">After the configuration is complete, you can view the agency configurations of an account or IAM user on the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1542514782712">Settings</strong> page. </p>
+</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li16660165611718">After configuring the username and agency, select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b04136444251">I have read and agree to the ModelArts Service Statement</strong> and click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b198971428172617">Agree</strong>.<p id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_p1010863219487">After the configuration is complete, you can view the agency configurations of an account or IAM user on the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1542514782712">Settings</strong> page.</p>
 </li></ol>
 </div>
-<div class="section" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"><a name="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"></a><a name="en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"></a><h4 class="sectiontitle">Automatically Creating an IAM Agency</h4><ul id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_ul20301022104513"><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li163042274515"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b15547121613219">Agency Name</strong>: For a cloud account, the agency name is <span class="parmvalue" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_parmvalue1017785818394"><b>modelarts_agency</b></span>. For an IAM user, the agency name is <span class="parmvalue" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_parmvalue165321902403"><b>ma_agency_&lt;IAM username&gt;</b></span>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li330192244515"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b264711611334">Agency Type</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b11371829123316">Cloud service</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li203062274518"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b149503410330">Cloud Service</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b23263573336">ModelArts</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li530182214453"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b799218471355">Validity Period</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1464905013515">Unlimited</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li1030822194510"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b745122225418">Permissions</strong>: The <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b889083635418">ModelArts CommonOperations</strong>, <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b4517124219547">OBS OperateAccess</strong>, and <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b7306114619544">Tenant Administrator</strong> (required for using other dependent services) permissions are automatically added for this agency to use all ModelArts functions. </li></ul>
+<div class="section" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"><a name="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"></a><a name="en-us_topic_0284258827_en-us_topic_0256240291_section19256347172519"></a><h4 class="sectiontitle">Automatically Creating an IAM Agency</h4><ul id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_ul20301022104513"><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li163042274515"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b15547121613219">Agency Name</strong>: For a cloud account, the agency name is <span class="parmvalue" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_parmvalue1017785818394"><b>modelarts_agency</b></span>. For an IAM user, the agency name is <span class="parmvalue" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_parmvalue165321902403"><b>ma_agency_&lt;IAM username&gt;</b></span>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li330192244515"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b264711611334">Agency Type</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b11371829123316">Cloud service</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li203062274518"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b149503410330">Cloud Service</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b23263573336">ModelArts</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li530182214453"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b799218471355">Validity Period</strong>: Select <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1464905013515">Unlimited</strong>.</li><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li1030822194510"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b745122225418">Permissions</strong>: The <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b889083635418">ModelArts CommonOperations</strong>, <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b4517124219547">OBS OperateAccess</strong>, and <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b7306114619544">Tenant Administrator</strong> (required for using other dependent services) permissions are automatically added for this agency to use all ModelArts functions.</li></ul>
 </div>
 <div class="section" id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_section15480411498"><h4 class="sectiontitle">Deleting Authorizations</h4><p id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_p16493812895">To better manage your authorization, you can delete the authorization of an IAM user or delete the authorizations of all users in batches.</p>
 <ul id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_ul15617115811184"><li id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_li161865819188"><strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b156681719306">Deleting the authorization of a user</strong><p id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_p63681037161918">On the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b114931834119">Settings</strong> page, the authorizations configured for IAM users under the current account are displayed. You can click <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b1931371114711">Delete</strong> in the <strong id="modelarts_08_0007__en-us_topic_0284258827_en-us_topic_0256240291_b2968161313713">Operation</strong> column to delete the authorization of a user. After the deletion takes effect, the user cannot use ModelArts functions.</p>

docs/modelarts/umn/modelarts_21_0003.html

@@ -8,7 +8,7 @@
 </div>
 <div class="section" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_section158944528285"><h4 class="sectiontitle">Requirements on Datasets</h4><ul id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_ul123600579125"><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li193610572124">The name of files in a dataset cannot contain Chinese characters, plus signs (+), spaces, or tabs.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li1736219578121">Ensure that no damaged image exists. The supported image formats include JPG, JPEG, BMP, and PNG.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li036285781212">Do not store data of different projects in the same dataset.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li336235741220">Prepare sufficient data and balance each class of data. To achieve better results, prepare at least 100 images of each class in a training set for image classification.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li13362857101211">To ensure the prediction accuracy of models, the training samples must be similar to the actual application scenarios.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li7362135741216">To ensure the generalization capability of models, datasets should cover all possible scenarios.</li></ul>
 </div>
-<div class="section" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_section8308141595310"><h4 class="sectiontitle">Requirements for Files Uploaded to OBS</h4><ul id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_ul133087155538"><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li12808450131514">If you do not need to upload training data in advance, create an empty folder to store files generated in the future, for example, <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath6742164318619"><b>/bucketName/data-cat</b></span>.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li2979741167">If you need to upload images to be labeled in advance, create an empty folder and save the images in the folder. An example of the image directory structure is <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath3979114141610"><b>/bucketName/data-cat/cat.jpg</b></span>.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li19308115205310">If you want to upload labeled images to the OBS bucket, upload them according to the following specifications:<ul id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_ul1308015205320"><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li774652717318">The dataset for image classification requires storing 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 is <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath16632143145312"><b>10.jpg</b></span>, the name of the label file must be <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath12479844155315"><b>10.txt</b></span>.<div class="p" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_p251410553285">Example of data files:<pre class="screen" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_screen10796103149">├─&lt;dataset-import-path&gt;
+<div class="section" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_section8308141595310"><h4 class="sectiontitle">Requirements for Files Uploaded to OBS</h4><ul id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_ul133087155538"><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li12808450131514">If you do not need to upload training data in advance, create an empty folder to store files generated in the future, for example, <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath6742164318619"><b>/bucketName/data-cat</b></span>.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li2979741167">If you need to upload images to be labeled in advance, create an empty folder and save the images in the folder. An example of the image directory structure is <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath3979114141610"><b>/bucketName/data-cat/cat.jpg</b></span>.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li19308115205310">If you want to upload labeled images to the OBS bucket, upload them according to the following specifications:<ul id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_ul1308015205320"><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li774652717318">The dataset for image classification requires storing 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 is <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath16632143145312"><b>10.jpg</b></span>, the name of the label file must be <span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath12479844155315"><b>10.txt</b></span>.<div class="p" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_p251410553285">Example of data files:<pre class="screen" id="modelarts_21_0003__screen1857612583569">├─&lt;dataset-import-path&gt;
       │      10.jpg
       │      10.txt    
       │      11.jpg 
@@ -17,7 +17,7 @@
       │      12.txt</pre>
 </div>
 </li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li1747821119592">Images in JPG, JPEG, PNG, and BMP formats are supported. When uploading images on the ModelArts management console, ensure that the size of an image does not exceed 5 MB and the total size of images to be uploaded in one attempt does not exceed 8 MB. If the data volume is large, use OBS Browser+ to upload images.</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li7168442469">A label name can contain a maximum of 32 characters, including Chinese characters, letters, digits, hyphens (-), and underscores (_).</li><li id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_li030671585320">Image classification label file (<span class="filepath" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_filepath14646751205811"><b>.txt</b></span>) rule:<p id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_p1130641513538">Each row contains only one label.</p>
-<pre class="screen" id="modelarts_21_0003__en-us_topic_0284258832_en-us_topic_0169446152_screen664210306592">cat
+<pre class="screen" id="modelarts_21_0003__screen442991512579">cat
 dog
 ...</pre>
 </li></ul>

docs/modelarts/umn/modelarts_21_0004.html

@@ -46,7 +46,7 @@
 </tr>
 <tr id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_row1094165617331"><td class="cellrowborder" valign="top" width="19.869999999999997%" headers="mcps1.3.2.2.3.2.2.3.1.1 "><p id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_p16720132684715">Label Set</p>
 </td>
-<td class="cellrowborder" valign="top" width="80.13%" headers="mcps1.3.2.2.3.2.2.3.1.2 "><ul id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_ul1050710321403"><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li165086322011"><strong id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_b1298114512011">Label Name</strong>: Enter a label name. The label name can contain only Chinese characters, letters, digits, underscores (_), and hyphens (-), which contains 1 to 32 characters.</li><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li750833218019"><span class="parmname" id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_parmname1115115586711"><b>Add Label</b></span>: Click <span class="parmname" id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_parmname147471328810"><b>Add Label</b></span> to add one or more labels.</li><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017">Set the label color: You need to set label colors for object detection datasets, but you do not need to set label colors for image classification datasets. Select a color from the color palette on the right of a label, or enter the hexadecimal color code to set the color.<p id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_p1755418011316"><a name="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017"></a><a name="en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017"></a></p>
+<td class="cellrowborder" valign="top" width="80.13%" headers="mcps1.3.2.2.3.2.2.3.1.2 "><ul id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_ul1050710321403"><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li165086322011"><strong id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_b1298114512011">Label Name</strong>: Enter a label name. The label name can contain only Chinese characters, letters, digits, underscores (_), and hyphens (-), which contains 1 to <span id="modelarts_21_0004__ph1841692418541">64</span>characters.</li><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li750833218019"><span class="parmname" id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_parmname1115115586711"><b>Add Label</b></span>: Click <span class="parmname" id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_parmname147471328810"><b>Add Label</b></span> to add one or more labels.</li><li id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017">Set the label color: You need to set label colors for object detection datasets, but you do not need to set label colors for image classification datasets. Select a color from the color palette on the right of a label, or enter the hexadecimal color code to set the color.<p id="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_p1755418011316"><a name="modelarts_21_0004__en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017"></a><a name="en-us_topic_0284258833_en-us_topic_0169446153_li18508203210017"></a></p>
 </li></ul>
 </td>
 </tr>

docs/modelarts/umn/modelarts_21_0006.html

@@ -51,10 +51,10 @@
 <tr id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_row138512264422"><td class="cellrowborder" valign="top" width="21.000000000000004%" headers="mcps1.3.2.2.2.6.2.4.1.1 "><p id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_p9386172624216">Instance Flavor</p>
 </td>
 <td class="cellrowborder" valign="top" width="57.00000000000001%" headers="mcps1.3.2.2.2.6.2.4.1.2 "><p id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_p913565015454">Select the resource specifications used for training. By default, the following specifications are supported:</p>
-<ul id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_ul39847524619"><li id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_li12984125114619"><span class="parmname" id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_parmname799064203417"><b>Compute-intensive 1 instance (GPU)</b></span>: This flavor is billed on a pay-per-use basis.</li></ul>
+<ul id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_ul39847524619"><li id="modelarts_21_0006__li2985185135513"><span class="parmname" id="modelarts_21_0006__parmname1111919545558"><b>Compute-intensive 1 instance (CPU)</b></span></li></ul>
 <p id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_p17822165817418">The compute flavors are for reference only. Obtain the flavors on the management console.</p>
 </td>
-<td class="cellrowborder" valign="top" width="22.000000000000004%" headers="mcps1.3.2.2.2.6.2.4.1.3 "><p id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_p10386162614212"><strong id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_b981218644517">ExeML (GPU)</strong></p>
+<td class="cellrowborder" valign="top" width="22.000000000000004%" headers="mcps1.3.2.2.2.6.2.4.1.3 "><p id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_p10386162614212"><strong id="modelarts_21_0006__en-us_topic_0284258835_en-us_topic_0169446155_b981218644517">ExeML (CPU)</strong></p>
 </td>
 </tr>
 </tbody>

docs/modelarts/umn/modelarts_21_0009.html

@@ -9,7 +9,7 @@
 <div class="section" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_section158944528285"><h4 class="sectiontitle">Requirements on Datasets</h4><ul id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_ul123600579125"><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li193610572124">The name of files in a dataset cannot contain Chinese characters, plus signs (+), spaces, or tabs.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li1736219578121">Ensure that no damaged image exists. The supported image formats include JPG, JPEG, BMP, and PNG.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li036285781212">Do not store data of different projects in the same dataset.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li13362857101211">To ensure the prediction accuracy of models, the training samples must be similar to the actual application scenarios.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_en-us_topic_0169446158_en-us_topic_0169446152_li7362135741216">To ensure the generalization capability of models, datasets should cover all possible scenarios.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_li1406171111918">In an object detection dataset, if the coordinates of the bounding box exceed the boundaries of an image, the image cannot be identified as a labeled image.</li></ul>
 </div>
 <div class="section" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_en-us_topic_0169446152_section8308141595310"><h4 class="sectiontitle">Requirements for Files Uploaded to OBS</h4><ul id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_ul366913119228"><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_li12808450131514">If you do not need to upload training data in advance, create an empty folder to store files generated in the future, for example, <span class="filepath" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_filepath58551116191112"><b>/bucketName/data-cat</b></span>.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_li2979741167">If you need to upload images to be labeled in advance, create an empty folder and save the images in the folder. An example of the image directory structure is <span class="filepath" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_filepath3979114141610"><b>/bucketName/data-cat/cat.jpg</b></span>.</li><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_li19154546102215">If you want to upload labeled images to the OBS bucket, upload them according to the following specifications:<ul id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_ul16340584619"><li id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_li46348574614">The dataset for object detection requires storing 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 is <span class="filepath" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_filepath16632143145312"><b>IMG_20180919_114745.jpg</b></span>, the name of the label file must be <span class="filepath" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_filepath12479844155315"><b>IMG_20180919_114745.xml</b></span>.<p id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_p199324710527">The label files for object detection must be in PASCAL VOC format. For details about the format, see <a href="#modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_table18220153119617">Table 1</a>.</p>
-<div class="p" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_p1747810389526">Example of data files:<pre class="screen" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_screen1880254115215">├─&lt;dataset-import-path&gt; 
+<div class="p" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_p1747810389526">Example of data files:<pre class="screen" id="modelarts_21_0009__screen3763134315579">├─&lt;dataset-import-path&gt; 
       │      IMG_20180919_114732.jpg 
       │      IMG_20180919_114732.xml 
       │      IMG_20180919_114745.jpg 
@@ -90,7 +90,7 @@
 </tbody>
 </table>
 </div>
-<div class="p" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_p10221531966">Example of the label file in KITTI format:<pre class="screen" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_screen8221531362">&lt;annotation&gt;
+<div class="p" id="modelarts_21_0009__en-us_topic_0284258838_en-us_topic_0169446158_p10221531966">Example of the label file in KITTI format:<pre class="screen" id="modelarts_21_0009__screen1228760155820">&lt;annotation&gt;
    &lt;folder&gt;test_data&lt;/folder&gt;
    &lt;filename&gt;260730932.jpg&lt;/filename&gt;
    &lt;size&gt;

docs/modelarts/umn/modelarts_21_0010.html

@@ -46,7 +46,7 @@
 </tr>
 <tr id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_row1094165617331"><td class="cellrowborder" valign="top" width="19.869999999999997%" headers="mcps1.3.2.2.3.2.2.3.1.1 "><p id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_p16720132684715">Label Set</p>
 </td>
-<td class="cellrowborder" valign="top" width="80.13%" headers="mcps1.3.2.2.3.2.2.3.1.2 "><ul id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_ul1050710321403"><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li165086322011"><strong id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_b1298114512011">Label Name</strong>: Enter a label name. The label name can contain only Chinese characters, letters, digits, underscores (_), and hyphens (-), which contains 1 to 32 characters.</li><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li750833218019"><span class="parmname" id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_parmname1115115586711"><b>Add Label</b></span>: Click <span class="parmname" id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_parmname147471328810"><b>Add Label</b></span> to add one or more labels.</li><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017">Set the label color: You need to set label colors for object detection datasets, but you do not need to set label colors for image classification datasets. Select a color from the color palette on the right of a label, or enter the hexadecimal color code to set the color.<p id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_p1755418011316"><a name="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017"></a><a name="en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017"></a></p>
+<td class="cellrowborder" valign="top" width="80.13%" headers="mcps1.3.2.2.3.2.2.3.1.2 "><ul id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_ul1050710321403"><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li165086322011"><strong id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_b1298114512011">Label Name</strong>: Enter a label name. The label name can contain only Chinese characters, letters, digits, underscores (_), and hyphens (-), which contains 1 to <span id="modelarts_21_0010__ph8275191410569">64</span> characters.</li><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li750833218019"><span class="parmname" id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_parmname1115115586711"><b>Add Label</b></span>: Click <span class="parmname" id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_parmname147471328810"><b>Add Label</b></span> to add one or more labels.</li><li id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017">Set the label color: You need to set label colors for object detection datasets, but you do not need to set label colors for image classification datasets. Select a color from the color palette on the right of a label, or enter the hexadecimal color code to set the color.<p id="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_p1755418011316"><a name="modelarts_21_0010__en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017"></a><a name="en-us_topic_0284258839_en-us_topic_0169446159_en-us_topic_0169446153_li18508203210017"></a></p>
 </li></ul>
 </td>
 </tr>

docs/modelarts/umn/modelarts_21_0012.html

@@ -51,10 +51,10 @@
 <tr id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_row138512264422"><td class="cellrowborder" valign="top" width="20%" headers="mcps1.3.2.2.2.6.2.4.1.1 "><p id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_p9386172624216">Instance Flavor</p>
 </td>
 <td class="cellrowborder" valign="top" width="59%" headers="mcps1.3.2.2.2.6.2.4.1.2 "><p id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_p913565015454">Select the resource specifications used for training. By default, the following specifications are supported:</p>
-<ul id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_ul39847524619"><li id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_li12984125114619"><span class="parmname" id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_parmname19637323203311"><b>Compute-intensive 1 instance (GPU)</b></span>: This flavor is billed on a pay-per-use basis.</li></ul>
+<ul id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_ul39847524619"><li id="modelarts_21_0012__li15718131125917"><span class="parmname" id="modelarts_21_0012__parmname1234513317592"><b>Compute-intensive 1 instance (CPU)</b></span></li></ul>
 <p id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_p1019151414233">The compute flavors are for reference only. Obtain the flavors on the management console.</p>
 </td>
-<td class="cellrowborder" valign="top" width="21%" headers="mcps1.3.2.2.2.6.2.4.1.3 "><p id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_p10386162614212"><strong id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_b6657164344618">ExeML (GPU)</strong></p>
+<td class="cellrowborder" valign="top" width="21%" headers="mcps1.3.2.2.2.6.2.4.1.3 "><p id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_p10386162614212"><strong id="modelarts_21_0012__en-us_topic_0284258841_en-us_topic_0169446261_b6657164344618">ExeML (CPU)</strong></p>
 </td>
 </tr>
 </tbody>

docs/modelarts/umn/modelarts_21_0016.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">Creating a Project</h1>
 <div id="body0000001251088334"><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p454323512370">ModelArts ExeML supports image classification, object detection, and predictive analytics projects. You can create any of them based on your needs. Perform the following operations to create an ExeML project.</p>
-<div class="section" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_section2361531133712"><h4 class="sectiontitle">Procedure</h4><ol id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_ol121861846192412"><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li190485518615">Log in to the ModelArts management console. In the left navigation pane, click <span class="parmname" id="modelarts_21_0016__parmname1395202883315"><b>ExeML</b></span>. The <span class="wintitle" id="modelarts_21_0016__wintitle1495216288333"><b>ExeML</b></span> page is displayed. </li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li1118634612413">Click <span class="parmname" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_parmname871519178378"><b>Create Project</b></span> in the box of your desired project. The page for creating an ExeML project is displayed. </li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li226801934314">Enter a project name and set <span class="parmname" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_parmname3550138164115"><b>Training Data</b></span> to the OBS path of the training data. A data file must be specified in the path.
+<div class="section" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_section2361531133712"><h4 class="sectiontitle">Procedure</h4><ol id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_ol121861846192412"><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li190485518615">Log in to the ModelArts management console. In the left navigation pane, click <span class="parmname" id="modelarts_21_0016__parmname1395202883315"><b>ExeML</b></span>. The <span class="wintitle" id="modelarts_21_0016__wintitle1495216288333"><b>ExeML</b></span> page is displayed. </li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li1118634612413">Click <span class="parmname" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_parmname871519178378"><b>Create Project</b></span> in the box of your desired project. The page for creating an ExeML project is displayed.</li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li226801934314">Enter a project name and set <span class="parmname" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_parmname3550138164115"><b>Training Data</b></span> to the OBS path of the training data. A data file must be specified in the path.
 <div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_table14961618163816" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Parameter description</caption><thead align="left"><tr id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_row11979189381"><th align="left" class="cellrowborder" valign="top" width="19.86%" id="mcps1.3.2.2.3.2.2.3.1.1"><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p436182517384">Parameter</p>
 </th>
 <th align="left" class="cellrowborder" valign="top" width="80.14%" id="mcps1.3.2.2.3.2.2.3.1.2"><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p1497111823813">Description</p>
@@ -12,7 +12,7 @@
 <tbody><tr id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_row1397181810385"><td class="cellrowborder" valign="top" width="19.86%" headers="mcps1.3.2.2.3.2.2.3.1.1 "><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p153211016173919">Name</p>
 </td>
 <td class="cellrowborder" valign="top" width="80.14%" headers="mcps1.3.2.2.3.2.2.3.1.2 "><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p174771543154614">Name of an ExeML project</p>
-<ul id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_ul1548720581506"><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li10487185818013">Enter a maximum of 20 characters. Only digits, letters, underscores (_), and hyphens (-) are allowed. This parameter is mandatory.</li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li174611121613">The name must start with a letter.</li></ul>
+<ul id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_ul1548720581506"><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li10487185818013">Enter a maximum of <span id="modelarts_21_0016__ph1514743111571">32</span> characters. Only digits, letters, underscores (_), and hyphens (-) are allowed. This parameter is mandatory.</li><li id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_li174611121613">The name must start with a letter.</li></ul>
 </td>
 </tr>
 <tr id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_row8987181381"><td class="cellrowborder" valign="top" width="19.86%" headers="mcps1.3.2.2.3.2.2.3.1.1 "><p id="modelarts_21_0016__en-us_topic_0169446265_en-us_topic_0169446153_p7324171603910">Training Data</p>

docs/modelarts/umn/modelarts_21_0019.html

@@ -8,43 +8,24 @@
 </li><li id="modelarts_21_0019__en-us_topic_0169446268_en-us_topic_0169446156_li11245345209">After the model is deployed, view the model deployment status on the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle5329114402917"><b>Service Deployment</b></span> page.<p id="modelarts_21_0019__en-us_topic_0169446268_en-us_topic_0169446156_p139485445203">The deployment takes a certain period of time. If the status in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname6741165118302"><b>Version Manager</b></span> pane changes from <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname49581321312"><b>Deploying</b></span> to <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1960319143117"><b>Running</b></span>, the deployment is complete.</p>
 </li></ol>
 </div>
-<div class="section" id="modelarts_21_0019__en-us_topic_0169446268_section535154714515"><h4 class="sectiontitle">Testing a Service</h4><ul id="modelarts_21_0019__en-us_topic_0169446268_ul81081912313"><li id="modelarts_21_0019__en-us_topic_0169446268_li6108811637">On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle194292181965"><b>Service Deployment</b></span> page, select a service type. For example, on the ExeML page, the predictive analytics model is deployed as a real-time service by default. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle9431161815616"><b>Real-Time Services</b></span> page, click <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname14338189617"><b>Prediction</b></span> in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname743581814618"><b>Operation</b></span> column of the target service to perform a service test. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0062.html" target="_blank" rel="noopener noreferrer">Testing a Service</a>.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li144461334311">You can also use code to test a service. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0063.html" target="_blank" rel="noopener noreferrer">Accessing a Real-Time Service</a>.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li3528165337">The following describes the procedure for performing a service test after the predictive analytics model is deployed as a service on the ExeML page.<ol id="modelarts_21_0019__en-us_topic_0169446268_ol939671815523"><li id="modelarts_21_0019__en-us_topic_0169446268_li9771161119539">After the model is deployed, you can test the model using code. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle1365144917715"><b>ExeML</b></span> page, click the target project, go to the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle17819167489"><b>Deployment Online</b></span> tab page, select the service version in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1735192713812"><b>Running</b></span> state, and enter the code in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1978181251512"><b>Code</b></span> area.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li1786641311616">Click <span class="uicontrol" id="modelarts_21_0019__en-us_topic_0169446268_uicontrol1181303111013"><b>Prediction</b></span> to perform the test. After the prediction is complete, the result is displayed in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname88392538102"><b>Return Result</b></span> area on the right. If the model accuracy does not meet your expectation, train and deploy the model again on the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname126425481116"><b>Label Data</b></span> tab page. If you are satisfied with the model prediction result, call the API to access the real-time service as prompted. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0063.html" target="_blank" rel="noopener noreferrer">Accessing a Real-Time Service</a>.<ul id="modelarts_21_0019__en-us_topic_0169446268_ul185518412710"><li id="modelarts_21_0019__en-us_topic_0169446268_li1485517417713"><span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname3902454163210"><b>attr_1</b></span> to <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname89991358113216"><b>attr_7</b></span> indicate the input data. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle873494113319"><b>Label Data</b></span> tab page, the selected label column is <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1274418913331"><b>attr_7</b></span>, that is, <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname475416137330"><b>attr_7</b></span> is the target column to be predicted. The value of <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1233913205336"><b>attr_7</b></span> can be set to any value or left blank, which does not affect the prediction result.<div class="codecoloring" codetype="Python" id="modelarts_21_0019__en-us_topic_0169446268_screen058516121173"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre> 1
+<div class="section" id="modelarts_21_0019__en-us_topic_0169446268_section535154714515"><h4 class="sectiontitle">Testing a Service</h4><ul id="modelarts_21_0019__en-us_topic_0169446268_ul81081912313"><li id="modelarts_21_0019__en-us_topic_0169446268_li6108811637">On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle194292181965"><b>Service Deployment</b></span> page, select a service type. For example, on the ExeML page, the predictive analytics model is deployed as a real-time service by default. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle9431161815616"><b>Real-Time Services</b></span> page, click <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname14338189617"><b>Prediction</b></span> in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname743581814618"><b>Operation</b></span> column of the target service to perform a service test. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0062.html" target="_blank" rel="noopener noreferrer">Testing a Service</a>.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li144461334311">You can also use code to test a service. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0063.html" target="_blank" rel="noopener noreferrer">Accessing a Real-Time Service</a>.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li3528165337">The following describes the procedure for performing a service test after the predictive analytics model is deployed as a service on the ExeML page.<ol id="modelarts_21_0019__en-us_topic_0169446268_ol939671815523"><li id="modelarts_21_0019__en-us_topic_0169446268_li9771161119539">After the model is deployed, you can test the model using code. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle1365144917715"><b>ExeML</b></span> page, click the target project, go to the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle17819167489"><b>Deployment Online</b></span> tab page, select the service version in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1735192713812"><b>Running</b></span> state, and enter the code in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1978181251512"><b>Code</b></span> area.</li><li id="modelarts_21_0019__en-us_topic_0169446268_li1786641311616">Click <span class="uicontrol" id="modelarts_21_0019__en-us_topic_0169446268_uicontrol1181303111013"><b>Prediction</b></span> to perform the test. After the prediction is complete, the result is displayed in the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname88392538102"><b>Return Result</b></span> area on the right. If the model accuracy does not meet your expectation, train and deploy the model again on the <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname126425481116"><b>Label Data</b></span> tab page. If you are satisfied with the model prediction result, call the API to access the real-time service as prompted. For details, see <a href="https://docs.otc.t-systems.com/en-us/usermanual/modelarts/modelarts_23_0063.html" target="_blank" rel="noopener noreferrer">Accessing a Real-Time Service</a>.<ul id="modelarts_21_0019__en-us_topic_0169446268_ul185518412710"><li id="modelarts_21_0019__en-us_topic_0169446268_li1485517417713"><span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname3902454163210"><b>attr_1</b></span> to <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname89991358113216"><b>attr_7</b></span> indicate the input data. On the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle873494113319"><b>Label Data</b></span> tab page, the selected label column is <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1274418913331"><b>attr_7</b></span>, that is, <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname475416137330"><b>attr_7</b></span> is the target column to be predicted. The value of <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1233913205336"><b>attr_7</b></span> can be set to any value or left blank, which does not affect the prediction result.<pre class="screen" id="modelarts_21_0019__screen4768104625817">{
- 2
+  "data": 
- 3
+  {
- 4
+    "count": 1,
- 5
+    "req_data": 
- 6
+	[
- 7
+      {
- 8
+        "attr_1": "58",
- 9
+        "attr_2": "management",
-10
+        "attr_3": "married",
-11
+        "attr_4": "tertiary",
-12
+        "attr_5": "yes",
-13
+        "attr_6": "no",
-14
+        "attr_7": ""
-15
+      }
-16
+    ]
-17
+  }
-18</pre></div></td><td class="code"><div class="highlight"><pre><span></span><span class="p">{</span>
+}</pre>
-  <span class="s2">&quot;data&quot;</span><span class="p">:</span> 
-  <span class="p">{</span>
-    <span class="s2">&quot;count&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
-    <span class="s2">&quot;req_data&quot;</span><span class="p">:</span> 
-	<span class="p">[</span>
-      <span class="p">{</span>
-        <span class="s2">&quot;attr_1&quot;</span><span class="p">:</span> <span class="s2">&quot;58&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_2&quot;</span><span class="p">:</span> <span class="s2">&quot;management&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_3&quot;</span><span class="p">:</span> <span class="s2">&quot;married&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_4&quot;</span><span class="p">:</span> <span class="s2">&quot;tertiary&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_5&quot;</span><span class="p">:</span> <span class="s2">&quot;yes&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_6&quot;</span><span class="p">:</span> <span class="s2">&quot;no&quot;</span><span class="p">,</span>
-        <span class="s2">&quot;attr_7&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span>
-      <span class="p">}</span>
-    <span class="p">]</span>
-  <span class="p">}</span>
-<span class="p">}</span>
-</pre></div>
-</td></tr></table></div>
 </li><li id="modelarts_21_0019__en-us_topic_0169446268_li1825064971513">In the preceding code snippet, <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname19153536193317"><b>predictioncol</b></span> is the inference result of label column <span class="parmname" id="modelarts_21_0019__en-us_topic_0169446268_parmname1371513402333"><b>attr_7</b></span>.</li></ul>
 <div class="note" id="modelarts_21_0019__en-us_topic_0169446268_note7312157124013"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><p id="modelarts_21_0019__en-us_topic_0169446268_p83126764019">A running real-time service keeps consuming the resources. If you do not need to use the real-time service, you are advised to click <span class="uicontrol" id="modelarts_21_0019__en-us_topic_0169446268_uicontrol4103184773210"><b>Stop</b></span> in the <span class="wintitle" id="modelarts_21_0019__en-us_topic_0169446268_wintitle111038478324"><b>Version Manager</b></span> pane to stop the service. If you want to use the service again, click <span class="uicontrol" id="modelarts_21_0019__en-us_topic_0169446268_uicontrol1438102414236"><b>Start</b></span>.</p>
 </div></div>

docs/modelarts/umn/modelarts_21_0057.html

@@ -1,7 +1,7 @@
 <a name="modelarts_21_0057"></a><a name="modelarts_21_0057"></a>
 
 <h1 class="topictitle1">Does ModelArts Support Multiple Projects?</h1>
-<div id="body0000001278593877"><p id="modelarts_21_0057__p185292015219">No. The current ModelArts version does not support multiple projects. Customers can only use it in the default eu-de project. </p>
+<div id="body0000001278593877"><p id="modelarts_21_0057__p185292015219">The current version supports multiple projects.</p>
 <p id="modelarts_21_0057__p15513193211451"></p>
 </div>
 <div>

docs/modelarts/umn/modelarts_21_0072.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">How Do I View GPU Usage on the Notebook?</h1>
 <div id="body0000001279857933"><p id="modelarts_21_0072__en-us_topic_0198037463_p6762184501911">If you select GPU when creating a notebook instance, perform the following operations to view GPU usage:</p>
-<ol id="modelarts_21_0072__en-us_topic_0198037463_ol1916513255134"><li id="modelarts_21_0072__en-us_topic_0198037463_li6683105013126">Log in to the ModelArts management console, and choose <span class="menucascade" id="modelarts_21_0072__en-us_topic_0198037463_menucascade17843196322"><b><span class="uicontrol" id="modelarts_21_0072__uicontrol138317198326">DevEnviron &gt; Notebooks</span></b></span>.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1521334591415">In the <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname11602721103211"><b>Operation</b></span> column of the target notebook instance in the notebook list, click <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname4603122163213"><b>Open</b></span> to go to the <span class="wintitle" id="modelarts_21_0072__en-us_topic_0198037463_wintitle1860492111325"><b>Jupyter</b></span> page.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1735811119618">On the <span class="wintitle" id="modelarts_21_0072__en-us_topic_0198037463_wintitle38451823173219"><b>Files</b></span> tab page of the <strong id="modelarts_21_0072__en-us_topic_0198037463_b1391193293213">Jupyter</strong> page, click <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname484672343210"><b>New</b></span> and select <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname9846423113220"><b>Terminal</b></span>. The <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname198471238324"><b>Terminal</b></span> page is displayed.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1564122461518">Run the following command to view GPU usage:<pre class="screen" id="modelarts_21_0072__en-us_topic_0198037463_screen217621774013">nvidia-smi</pre>
+<ol id="modelarts_21_0072__en-us_topic_0198037463_ol1916513255134"><li id="modelarts_21_0072__en-us_topic_0198037463_li6683105013126">Log in to the ModelArts management console, and choose <span class="menucascade" id="modelarts_21_0072__en-us_topic_0198037463_menucascade17843196322"><b><span class="uicontrol" id="modelarts_21_0072__uicontrol138317198326">DevEnviron &gt; Notebooks</span></b></span>.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1521334591415">In the <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname11602721103211"><b>Operation</b></span> column of the target notebook instance in the notebook list, click <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname4603122163213"><b>Open</b></span> to go to the <span class="wintitle" id="modelarts_21_0072__en-us_topic_0198037463_wintitle1860492111325"><b>Jupyter</b></span> page.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1735811119618">On the <span class="wintitle" id="modelarts_21_0072__en-us_topic_0198037463_wintitle38451823173219"><b>Files</b></span> tab page of the <strong id="modelarts_21_0072__en-us_topic_0198037463_b1391193293213">Jupyter</strong> page, click <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname484672343210"><b>New</b></span> and select <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname9846423113220"><b>Terminal</b></span>. The <span class="parmname" id="modelarts_21_0072__en-us_topic_0198037463_parmname198471238324"><b>Terminal</b></span> page is displayed.</li><li id="modelarts_21_0072__en-us_topic_0198037463_li1564122461518">Run the following command to view GPU usage:<pre class="screen" id="modelarts_21_0072__screen188691859259">nvidia-smi</pre>
 </li></ol>
 </div>
 <div>

docs/modelarts/umn/modelarts_21_0077.html

@@ -4,7 +4,7 @@
 <div id="body0000001279824501"><p id="modelarts_21_0077__en-us_topic_0198838770_p548115183417">Pay attention to the following when setting training parameters:</p>
 <ul id="modelarts_21_0077__en-us_topic_0198838770_ul15222135643318"><li id="modelarts_21_0077__en-us_topic_0198838770_li18387146144914">When setting running parameters for creating a training job, you only need to set the corresponding parameter names and values. See <a href="#modelarts_21_0077__en-us_topic_0198838770_fig1638314610491">Figure 1</a>.<div class="fignone" id="modelarts_21_0077__en-us_topic_0198838770_fig1638314610491"><a name="modelarts_21_0077__en-us_topic_0198838770_fig1638314610491"></a><a name="en-us_topic_0198838770_fig1638314610491"></a><span class="figcap"><b>Figure 1 </b>Setting running parameters</span><br><span><img id="modelarts_21_0077__en-us_topic_0198838770_image1944064817263" src="en-us_image_0000001235985012.png"></span></div>
 </li><li id="modelarts_21_0077__en-us_topic_0198838770_li1493031385918">If a parameter value is an OBS bucket path, use the path to the data. See <a href="#modelarts_21_0077__en-us_topic_0198838770_fig392681318596">Figure 2</a>.<div class="fignone" id="modelarts_21_0077__en-us_topic_0198838770_fig392681318596"><a name="modelarts_21_0077__en-us_topic_0198838770_fig392681318596"></a><a name="en-us_topic_0198838770_fig392681318596"></a><span class="figcap"><b>Figure 2 </b>Configuring an OBS path</span><br><span><img id="modelarts_21_0077__en-us_topic_0198838770_image65067337254" src="en-us_image_0000001235665800.png"></span></div>
-</li><li id="modelarts_21_0077__en-us_topic_0198838770_li1132272718180">When creating an OBS folder in code, you need to call a MoXing API as follows:<pre class="screen" id="modelarts_21_0077__en-us_topic_0198838770_screen11301434184">import moxing as mox
+</li><li id="modelarts_21_0077__en-us_topic_0198838770_li1132272718180">When creating an OBS folder in code, you need to call a MoXing API as follows:<pre class="screen" id="modelarts_21_0077__screen1597419391263">import moxing as mox
 mox.file.make_dirs('obs://bucket_name/sub_dir_0/sub_dir_1')</pre>
 </li></ul>
 </div>

docs/modelarts/umn/modelarts_21_0079.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">How Do I Check Whether Folder Copy Is Complete During Job Training?</h1>
 <div id="body0000001279665293"><p id="modelarts_21_0079__en-us_topic_0200343604_p235429583">In the script of the training job boot file, run the following commands to obtain the sizes of the to-be-copied and copied folders. Then determine whether folder copy is complete based on the command output.</p>
-<pre class="screen" id="modelarts_21_0079__en-us_topic_0200343604_screen58061415195915">import moxing as mox
+<pre class="screen" id="modelarts_21_0079__screen7340135732610">import moxing as mox
 mox.file.get_size('obs://bucket_name/obs_file',recursive=True)</pre>
 <p id="modelarts_21_0079__en-us_topic_0200343604_p1855172491214"><span class="parmvalue" id="modelarts_21_0079__en-us_topic_0200343604_parmvalue13550142491217"><b>get_size</b></span> indicates the size of the file or folder to be obtained. <span class="parmvalue" id="modelarts_21_0079__en-us_topic_0200343604_parmvalue1755112451214"><b>recursive=True</b></span> indicates that the type is folder. <span class="parmvalue" id="modelarts_21_0079__en-us_topic_0200343604_parmvalue7551202415121"><b>True</b></span> indicates that the type is folder, and <span class="parmvalue" id="modelarts_21_0079__en-us_topic_0200343604_parmvalue16551122414122"><b>False</b></span> indicates that the type is file.</p>
 <p id="modelarts_21_0079__en-us_topic_0200343604_p1271319499482">If the command output is consistent, the folder copy is complete. If the command output is inconsistent, the folder copy is not complete.</p>

docs/modelarts/umn/modelarts_21_0080.html

@@ -3,7 +3,7 @@
 <h1 class="topictitle1">How Do I Obtain Training Job Parameters from the Boot File of the Training Job?</h1>
 <div id="body0000001279984733"><p id="modelarts_21_0080__en-us_topic_0200386929_p142239014391">Training job parameters can be automatically generated in the background or manually entered by users. Perform the following operations to obtain training job parameters:</p>
 <ol id="modelarts_21_0080__en-us_topic_0200386929_ol6422945164118"><li id="modelarts_21_0080__en-us_topic_0200386929_li1442214452411">When a training job is created, <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue152033199404"><b>train_url</b></span> in the running parameters of the training job indicates a training output location, and <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue8210112704012"><b>data_url</b></span> indicates a data source, which is automatically generated in the background. The <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue209191426406"><b>test</b></span> parameter is manually entered.</li><li id="modelarts_21_0080__en-us_topic_0200386929_li20867185444120">After the training job is executed, you can click the job name in the training job list to view its details. You can obtain the parameter input mode from logs, as shown in <a href="#modelarts_21_0080__en-us_topic_0200386929_fig74929528456">Figure 1</a>.<div class="fignone" id="modelarts_21_0080__en-us_topic_0200386929_fig74929528456"><a name="modelarts_21_0080__en-us_topic_0200386929_fig74929528456"></a><a name="en-us_topic_0200386929_fig74929528456"></a><span class="figcap"><b>Figure 1 </b>Viewing logs</span><br><span><img id="modelarts_21_0080__en-us_topic_0200386929_image183416221441" src="en-us_image_0000001279825393.png"></span></div>
-</li><li id="modelarts_21_0080__en-us_topic_0200386929_li1449212523451">To obtain the values of <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue186245164415"><b>train_url</b></span>, <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue1686265114418"><b>data_url</b></span>, and <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue167781319104613"><b>test</b></span> during training, add the following code to the boot file of the training job:<pre class="screen" id="modelarts_21_0080__en-us_topic_0200386929_screen1533211718467">import argparse
+</li><li id="modelarts_21_0080__en-us_topic_0200386929_li1449212523451">To obtain the values of <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue186245164415"><b>train_url</b></span>, <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue1686265114418"><b>data_url</b></span>, and <span class="parmvalue" id="modelarts_21_0080__en-us_topic_0200386929_parmvalue167781319104613"><b>test</b></span> during training, add the following code to the boot file of the training job:<pre class="screen" id="modelarts_21_0080__screen1442515127270">import argparse
 parser = argparse.ArgumentParser()
 parser.add_argument('--data_url', type=str, default=None, help='test')
 parser.add_argument('--train_url', type=str, default=None, help='test')

docs/modelarts/umn/modelarts_21_0083.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">Only Three Valid Digits Are Retained in a Training Output Log. Can the Value of loss Be Changed?</h1>
 <div id="body0000001235824820"><p id="modelarts_21_0083__en-us_topic_0200392768_p1369712501911">In a training job, only three valid digits are retained in a training output log. When the value of <strong id="modelarts_21_0083__en-us_topic_0200392768_b184901439151515">loss</strong> is too small, the value is displayed as <strong id="modelarts_21_0083__en-us_topic_0200392768_b151111744171516">0.000</strong>. Log content is as follows:</p>
-<pre class="screen" id="modelarts_21_0083__en-us_topic_0200392768_screen9660648111020">INFO:tensorflow:global_step/sec: 0.382191
+<pre class="screen" id="modelarts_21_0083__screen5870825172716">INFO:tensorflow:global_step/sec: 0.382191
 INFO:tensorflow:step: 81600(global step: 81600) sample/sec: 12.098 loss: 0.000
 INFO:tensorflow:global_step/sec: 0.382876
 INFO:tensorflow:step: 81700(global step: 81700) sample/sec: 12.298 loss: 0.000</pre>

docs/modelarts/umn/modelarts_21_0084.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">Why Can't I Use os.system ('cd xxx') to Access the Corresponding Folder During Job Training?</h1>
 <div id="body0000001235984136"><p id="modelarts_21_0084__en-us_topic_0200392769_p5401866496">If you cannot access the corresponding folder by using <b><span class="cmdname" id="modelarts_21_0084__en-us_topic_0200392769_cmdname1821315107516">os.system('cd xxx')</span></b> in the boot script of the training job, you are advised to use the following method:</p>
-<pre class="screen" id="modelarts_21_0084__en-us_topic_0200392769_screen8664643653">import os
+<pre class="screen" id="modelarts_21_0084__screen974115407270">import os
 os.chdir('/home/work/user-job-dir/xxx')</pre>
 </div>
 <div>

docs/modelarts/umn/modelarts_21_0085.html

@@ -2,7 +2,7 @@
 
 <h1 class="topictitle1">How Do I Invoke a Shell Script in a Training Job to Execute the .sh File?</h1>
 <div id="body0000001279824505"><p id="modelarts_21_0085__en-us_topic_0200392419_p109071418171517">ModelArts enables you to invoke a shell script, and you can use Python to invoke <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath69074187152"><b>.sh</b></span>. The procedure is as follows:</p>
-<ol id="modelarts_21_0085__en-us_topic_0200392419_ol6632173210442"><li id="modelarts_21_0085__en-us_topic_0200392419_li263253244410">Upload the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath149021434572"><b>.sh</b></span> script to an OBS bucket. For example, upload the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath38685711475"><b>.sh</b></span> script to <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath1427612391710"><b>/bucket-name/code/test.sh</b></span>.</li><li id="modelarts_21_0085__en-us_topic_0200392419_li16279132235116">Create the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath92811022145118"><b>.py</b></span> file on a local PC, for example, <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath628162216519"><b>test.py</b></span>. The background automatically downloads the code directory to the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath203534819920"><b>/home/work/user-job-dir/</b></span> directory of the container. Therefore, you can invoke the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath959132117495"><b>.sh</b></span> file in the <span class="parmname" id="modelarts_21_0085__en-us_topic_0200392419_parmname19591102119498"><b>test.py</b></span> boot file as follows:<pre class="screen" id="modelarts_21_0085__en-us_topic_0200392419_screen45918211496">import os
+<ol id="modelarts_21_0085__en-us_topic_0200392419_ol6632173210442"><li id="modelarts_21_0085__en-us_topic_0200392419_li263253244410">Upload the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath149021434572"><b>.sh</b></span> script to an OBS bucket. For example, upload the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath38685711475"><b>.sh</b></span> script to <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath1427612391710"><b>/bucket-name/code/test.sh</b></span>.</li><li id="modelarts_21_0085__en-us_topic_0200392419_li16279132235116">Create the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath92811022145118"><b>.py</b></span> file on a local PC, for example, <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath628162216519"><b>test.py</b></span>. The background automatically downloads the code directory to the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath203534819920"><b>/home/work/user-job-dir/</b></span> directory of the container. Therefore, you can invoke the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath959132117495"><b>.sh</b></span> file in the <span class="parmname" id="modelarts_21_0085__en-us_topic_0200392419_parmname19591102119498"><b>test.py</b></span> boot file as follows:<pre class="screen" id="modelarts_21_0085__screen134361254172717">import os
 os.system('bash /home/work/user-job-dir/code/test.sh')</pre>
 </li><li id="modelarts_21_0085__en-us_topic_0200392419_li16333115295010">Upload <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath96181443175518"><b>test.py</b></span> to OBS. Then the file storage path is <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath192811722165117"><b>/bucket-name/code/test.py</b></span>.</li><li id="modelarts_21_0085__en-us_topic_0200392419_li310415765316">When creating a training job, set the code directory to <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath10104107145314"><b>/bucket-name/code/</b></span>, and the boot file directory to <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath9651531283"><b>/bucket-name/code/test.py</b></span>.</li></ol>
 <p id="modelarts_21_0085__en-us_topic_0200392419_p15607019125420">After the training job is created, you can use Python to invoke the <span class="filepath" id="modelarts_21_0085__en-us_topic_0200392419_filepath1957212569544"><b>.sh</b></span> file.</p>