forked from docs/doc-exports
Lai, Weijian
6aa966a79a
Reviewed-by: Pruthi, Vineet <vineet.pruthi@t-systems.com> Co-authored-by: Lai, Weijian <laiweijian4@huawei.com> Co-committed-by: Lai, Weijian <laiweijian4@huawei.com>
74 lines
15 KiB
HTML
74 lines
15 KiB
HTML
<a name="EN-US_TOPIC_0000002043022660"></a><a name="EN-US_TOPIC_0000002043022660"></a>
|
|
|
|
<h1 class="topictitle1">Creating an Auto Labeling Job</h1>
|
|
<div id="body0000001210404157"><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p19594032182313">In addition to manual labeling, ModelArts also provides the auto labeling function to quickly label data, reducing the labeling time by more than 70%. Auto labeling means learning and training are performed based on the labeled images and an existing model is used to quickly label the remaining images.</p>
|
|
<div class="section" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_section1054132514233"><h4 class="sectiontitle">Context</h4><ul id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_ul1593217319236"><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li1604195884318">Only labeling jobs of image classification and object detection types support auto labeling.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li293220310237">There are at least two types of labels in the labeling job for auto labeling, and each label has been added to at least five images.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li19932231152315">At least one unlabeled image must exist when you enable auto labeling.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li29321131172314">Before starting an auto labeling job, ensure that no auto labeling job is in progress.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li12384351409">Before starting an auto labeling job, ensure that the image data does not contain any RGBA four-channel images. These images will cause the job to fail. Delete them from the dataset if you find any.</li></ul>
|
|
</div>
|
|
<div class="section" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_section237785911185"><h4 class="sectiontitle">Starting an Auto Labeling Job</h4><ol id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_ol11378205918189"><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li05471715152412">Log in to the ModelArts management console. In the navigation pane on the left, choose <strong id="EN-US_TOPIC_0000002043022660__b24508341894858">Data Management</strong> > <span class="parmname" id="EN-US_TOPIC_0000002043022660__parmname39030167294858"><b>Label Data</b></span>. The <span class="wintitle" id="EN-US_TOPIC_0000002043022660__wintitle130919834194858"><b>Data Labeling</b></span> page is displayed.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li3463205211243">In the labeling job list, locate the target labeling job of the object detection or image classification type, and click <strong id="EN-US_TOPIC_0000002043022660__b193821305394858">Auto Labeling</strong> in the <strong id="EN-US_TOPIC_0000002043022660__b73003107894858">Operation</strong> column.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li19233182874814">On the <strong id="EN-US_TOPIC_0000002043022660__b91751744594858">Enable Auto Labeling</strong> page, select <strong id="EN-US_TOPIC_0000002043022660__b99882352794858">Active learning</strong> or <strong id="EN-US_TOPIC_0000002043022660__b145935259394858">Pre-labeling</strong>. For details, see <a href="#EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table898214481813">Table 1</a> and <a href="#EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table4748115061313">Table 2</a>.
|
|
<div class="tablenoborder"><a name="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table898214481813"></a><a name="en-us_topic_0000001185384429_en-us_topic_0209622045_table898214481813"></a><table cellpadding="4" cellspacing="0" summary="" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table898214481813" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Active learning</caption><thead align="left"><tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row898216482815"><th align="left" class="cellrowborder" valign="top" width="15.03%" id="mcps1.3.3.2.3.6.2.3.1.1"><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1298254816810">Parameter</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="84.97%" id="mcps1.3.3.2.3.6.2.3.1.2"><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1441851154219">Description</p>
|
|
</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody><tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row1281819611565"><td class="cellrowborder" valign="top" width="15.03%" headers="mcps1.3.3.2.3.6.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p78189610567">Auto Labeling Type</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="84.97%" headers="mcps1.3.3.2.3.6.2.3.1.2 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p16819116115611"><strong id="EN-US_TOPIC_0000002043022660__b101968524694858">Active learning</strong>: The system uses semi-supervised learning and hard example filtering to perform auto labeling, reducing manual labeling workload and helping you find hard examples.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row1498216481983"><td class="cellrowborder" valign="top" width="15.03%" headers="mcps1.3.3.2.3.6.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p139827481287">Algorithm Type</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="84.97%" headers="mcps1.3.3.2.3.6.2.3.1.2 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p104936439121">For a dataset of the image classification type, set the following parameters:</p>
|
|
<p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1398244812820"><strong id="EN-US_TOPIC_0000002043022660__b93788140694858">Fast</strong>: Use the labeled samples for training.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row7407154004112"><td class="cellrowborder" valign="top" width="15.03%" headers="mcps1.3.3.2.3.6.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1440794010414">Specifications</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="84.97%" headers="mcps1.3.3.2.3.6.2.3.1.2 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1840794010417">Resource specifications used by an auto labeling job. Only GPU specifications are supported, and the OBT is free of charge.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row225783717417"><td class="cellrowborder" valign="top" width="15.03%" headers="mcps1.3.3.2.3.6.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1257133712414">Compute Nodes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="84.97%" headers="mcps1.3.3.2.3.6.2.3.1.2 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p12573373418">The default value is <strong id="EN-US_TOPIC_0000002043022660__b81695410694858">1</strong>, indicating the single-node system mode. Only this parameter value is supported.</p>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
|
|
<div class="tablenoborder"><a name="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table4748115061313"></a><a name="en-us_topic_0000001185384429_en-us_topic_0209622045_table4748115061313"></a><table cellpadding="4" cellspacing="0" summary="" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_table4748115061313" frame="border" border="1" rules="all"><caption><b>Table 2 </b>Pre-labeling</caption><thead align="left"><tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row1974855081317"><th align="left" class="cellrowborder" valign="top" width="14.99%" id="mcps1.3.3.2.3.7.2.3.1.1"><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p17748950141319">Parameter</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="85.00999999999999%" id="mcps1.3.3.2.3.7.2.3.1.2"><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p47482504133">Description</p>
|
|
</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody><tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row1337895218585"><td class="cellrowborder" valign="top" width="14.99%" headers="mcps1.3.3.2.3.7.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p52196584587">Auto Labeling Type</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="85.00999999999999%" headers="mcps1.3.3.2.3.7.2.3.1.2 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p12379135214585"><strong id="EN-US_TOPIC_0000002043022660__b148368935594858">Pre-labeling</strong>: Select a model in the <strong id="EN-US_TOPIC_0000002043022660__b18704571594858">My AI Applications</strong> tab. Ensure that the model type matches the dataset labeling type. After the pre-labeling is complete, if the labeling result complies with the standard labeling format defined by the platform, the system filters hard examples. This step does not affect the pre-labeling result.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_row7748050151314"><td class="cellrowborder" valign="top" width="14.99%" headers="mcps1.3.3.2.3.7.2.3.1.1 "><p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p12748185051319">Model and Version</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="85.00999999999999%" headers="mcps1.3.3.2.3.7.2.3.1.2 "><ul id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_ul15863193332016"><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li386303322019"><strong id="EN-US_TOPIC_0000002043022660__b152121669494858">My AI Applications</strong>: Select a model as required. Click the drop-down arrow on the left of the target AI application and select a proper version. For details about how to import a model, see <a href="inference-modelarts-0004.html">Creating an AI Application</a></li></ul>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
<div class="note" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_note3922175611213"><img src="public_sys-resources/note_3.0-en-us.png"><span class="notetitle"> </span><div class="notebody"><ul id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_ul696580105820"><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li9165135155816">For labeling jobs of the object detection type, only rectangular boxes can be recognized and labeled when <span class="parmname" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_parmname01681056582"><b>Active learning</b></span> is selected.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li7966602588">If there are too many auto labeling jobs in the system, the jobs may be queued. As a result, the jobs are always in the labeling state. The jobs will run one after another in order.</li></ul>
|
|
</div></div>
|
|
<p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_p346816195416"></p>
|
|
</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li28153475110">After setting the parameters, click <strong id="EN-US_TOPIC_0000002043022660__b11573170094858">Submit</strong> to enable auto labeling.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li961312300527">In the labeling job list, click a labeling job name to go to the labeling job details page.</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_li557815310269">Click the <strong id="EN-US_TOPIC_0000002043022660__b95407538194858">To Be Confirmed</strong> tab to view the auto labeling progress.<p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_p657812311268">You can also enable auto labeling or view the auto labeling history in this tab.</p>
|
|
</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li119561935237">After auto labeling is complete, all the labeled images are displayed on the <strong id="EN-US_TOPIC_0000002043022660__b81775401294858">To Be Confirmed</strong> page.<ul id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_ul12714165219227"><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li6714752112217">Image classification labeling job<p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p9821111692"><a name="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li6714752112217"></a><a name="en-us_topic_0000001185384429_en-us_topic_0209622045_li6714752112217"></a>On the <span class="wintitle" id="EN-US_TOPIC_0000002043022660__wintitle176121652094858"><b>To Be Confirmed</b></span> page, check whether labels are correct, select the correctly labeled images, and click <strong id="EN-US_TOPIC_0000002043022660__b139809789594858">OK</strong> to confirm the auto labeling results. The confirmed image will be categorized to the <span class="parmname" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_parmname4379155941812"><b>Labeled</b></span> page.</p>
|
|
</li><li id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li21503682312">Object detection labeling job<p id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_p1377413179106"><a name="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_li21503682312"></a><a name="en-us_topic_0000001185384429_en-us_topic_0209622045_li21503682312"></a>On the <span class="wintitle" id="EN-US_TOPIC_0000002043022660__wintitle205591736694858"><b>To Be Confirmed</b></span> page, click images to view their labeling details and check whether labels and target bounding boxes are correct. For the correctly labeled images, click <strong id="EN-US_TOPIC_0000002043022660__b103927856794858">Labeled</strong> to confirm the auto labeling results. The confirmed image will be categorized to the <span class="parmname" id="EN-US_TOPIC_0000002043022660__en-us_topic_0000001185384429_en-us_topic_0209622045_parmname147747179104"><b>Labeled</b></span> page.</p>
|
|
</li></ul>
|
|
</li></ol>
|
|
</div>
|
|
</div>
|
|
<div>
|
|
<div class="familylinks">
|
|
<div class="parentlink"><strong>Parent topic:</strong> <a href="datalabel-modelarts_0018.html">Auto Labeling</a></div>
|
|
</div>
|
|
</div>
|
|
|