doc-exports/docs/modelarts/umn/datalabel-modelarts_0002.html
Lai, Weijian 6aa966a79a ModelArts UMN 24.3.0 version
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>
2024-11-02 09:04:52 +00:00

105 lines
9.3 KiB
HTML

<a name="EN-US_TOPIC_0000002079101737"></a><a name="EN-US_TOPIC_0000002079101737"></a>
<h1 class="topictitle1">Introduction to Data Labeling</h1>
<div id="body0000001193989297"><p id="EN-US_TOPIC_0000002079101737__p934525101412">Model training requires a large amount of labeled data. Therefore, before training a model, label data. ModelArts offers data labeling functions to assist with this process.</p>
<div class="section" id="EN-US_TOPIC_0000002079101737__section7692720192916"><h4 class="sectiontitle">Manual Labeling</h4><p id="EN-US_TOPIC_0000002079101737__p176100427206">Create a labeling job based on the dataset type. ModelArts supports the following types of labeling jobs:</p>
<ul id="EN-US_TOPIC_0000002079101737__ul8131611181911"><li id="EN-US_TOPIC_0000002079101737__li1013141110194">Image<ul id="EN-US_TOPIC_0000002079101737__ul11311151918"><li id="EN-US_TOPIC_0000002079101737__li5131011131918">Image classification: identifies a class of objects in images.</li><li id="EN-US_TOPIC_0000002079101737__li713611141916">Object detection: identifies the position and class of each object in an image.</li><li id="EN-US_TOPIC_0000002079101737__li872231914194">Image segmentation: segments an image into different areas based on objects in the image.</li></ul>
</li><li id="EN-US_TOPIC_0000002079101737__li2131211171915">Audio<ul id="EN-US_TOPIC_0000002079101737__ul51313111197"><li id="EN-US_TOPIC_0000002079101737__li19131011121919">Sound classification: classifies and identifies different sounds.</li><li id="EN-US_TOPIC_0000002079101737__li513181121916">Speech labeling: labels speech content.</li><li id="EN-US_TOPIC_0000002079101737__li121317118190">Speech paragraph labeling: segments and labels speech content.</li></ul>
</li><li id="EN-US_TOPIC_0000002079101737__li11131211111919">Text<ul id="EN-US_TOPIC_0000002079101737__ul16134118198"><li id="EN-US_TOPIC_0000002079101737__li111371161916">Text classification: assigns labels to text according to its content.</li><li id="EN-US_TOPIC_0000002079101737__li51371117196">Named entity recognition: assigns labels to named entities in text, such as time and locations.</li><li id="EN-US_TOPIC_0000002079101737__li61310113195">Text triplet: assigns labels to entity segments and entity relationships in the text.</li></ul>
</li><li id="EN-US_TOPIC_0000002079101737__li1159034212248">Video<p id="EN-US_TOPIC_0000002079101737__p24166716399"><a name="EN-US_TOPIC_0000002079101737__li1159034212248"></a><a name="li1159034212248"></a>Video labeling: identifies the position and class of each object in a video. Only the MP4 format is supported.</p>
</li></ul>
</div>
<div class="section" id="EN-US_TOPIC_0000002079101737__en-us_topic_0171496996_section10711124814415"><h4 class="sectiontitle">Dataset Functions</h4><p id="EN-US_TOPIC_0000002079101737__en-us_topic_0171496996_p2083514481545">Dataset functions vary depending on dataset types. For details, see <a href="#EN-US_TOPIC_0000002079101737__table475114812297">Table 1</a>.</p>
<div class="tablenoborder"><a name="EN-US_TOPIC_0000002079101737__table475114812297"></a><a name="table475114812297"></a><table cellpadding="4" cellspacing="0" summary="" id="EN-US_TOPIC_0000002079101737__table475114812297" width="100%" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Functions supported by different types of datasets</caption><thead align="left"><tr id="EN-US_TOPIC_0000002079101737__row127514842918"><th align="left" class="cellrowborder" valign="top" width="20.52%" id="mcps1.3.3.3.2.4.1.1"><p id="EN-US_TOPIC_0000002079101737__p1275168132913">Dataset Type</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="37.059999999999995%" id="mcps1.3.3.3.2.4.1.2"><p id="EN-US_TOPIC_0000002079101737__p11331215113511">Labeling Type</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="42.42%" id="mcps1.3.3.3.2.4.1.3"><p id="EN-US_TOPIC_0000002079101737__p1335515343119">Manual Labeling</p>
</th>
</tr>
</thead>
<tbody><tr id="EN-US_TOPIC_0000002079101737__row1475228202911"><td class="cellrowborder" rowspan="3" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p5752168112911">Image</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p15133121543510">Image classification</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p123561853153117">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row1812483210359"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p6125332173510">Object detection</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p181259327358">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row67675356352"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p13767193503513">Image segmentation</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p276793563512">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row18752148112912"><td class="cellrowborder" rowspan="3" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p1175219816293">Audio</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p2013314156358">Sound classification</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p1535616536318">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row266212973617"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p18663149173612">Speech Labeling</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p1166316913618">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row3359147113616"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p14359107113615">Speech Paragraph Labeling</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p13359179368">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row1875215822917"><td class="cellrowborder" rowspan="3" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p147522812920">Text</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p4133161593516">Text classification</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p43561553123117">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row6703348363"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p107003418365">Named entity recognition</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p1570173412364">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row354393163615"><td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p1554403183613">Text Triplet</p>
</td>
<td class="cellrowborder" valign="top" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p12544163112363">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row8752108102919"><td class="cellrowborder" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p127528802914">Videos</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p13133171516352">Video Labeling</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p8356153163115">Yes</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row775310862910"><td class="cellrowborder" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p9753389298">Free format</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p11133171514357">-</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p123564532319">-</p>
</td>
</tr>
<tr id="EN-US_TOPIC_0000002079101737__row137538862910"><td class="cellrowborder" valign="top" width="20.52%" headers="mcps1.3.3.3.2.4.1.1 "><p id="EN-US_TOPIC_0000002079101737__p1675318172912">Table</p>
</td>
<td class="cellrowborder" valign="top" width="37.059999999999995%" headers="mcps1.3.3.3.2.4.1.2 "><p id="EN-US_TOPIC_0000002079101737__p1613316154354">-</p>
</td>
<td class="cellrowborder" valign="top" width="42.42%" headers="mcps1.3.3.3.2.4.1.3 "><p id="EN-US_TOPIC_0000002079101737__p13356175313111">-</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div>
<div class="familylinks">
<div class="parentlink"><strong>Parent topic:</strong> <a href="modelarts_88_0148.html">Data Labeling</a></div>
</div>
</div>