vpruthi/doc-exports
1
0
Fork 0
forked from docs/doc-exports
Code Pull Requests Activity
doc-exports/docs/modelarts/sdk-ref/modelarts_04_0174.html
Lai, Weijian 1ad28cc33e modelarts sdk-ref first version 
Reviewed-by: Jiang, Beibei <beibei.jiang@t-systems.com>
Reviewed-by: Hasko, Vladimir <vladimir.hasko@t-systems.com>
Co-authored-by: Lai, Weijian <laiweijian4@huawei.com>
Co-committed-by: Lai, Weijian <laiweijian4@huawei.com>
2023-03-01 15:24:28 +00:00

318 lines
36 KiB
HTML
Raw Blame History

<a name="modelarts_04_0174"></a><a name="modelarts_04_0174"></a>
<h1 class="topictitle1">Creating a Training Job Configuration</h1>
<div id="body8662426"><div class="section" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_section35881040102516"><h4 class="sectiontitle">Sample Code</h4><p id="modelarts_04_0174__en-us_topic_0180094067_p58021924174412">In the ModelArts notebook instance, you do not need to enter authentication parameters for session authentication. For details about session authentication of other development environments, see <a href="modelarts_04_0123.html">Session Authentication</a>.</p>
<ul id="modelarts_04_0174__en-us_topic_0180094067_ul1115151011551"><li id="modelarts_04_0174__en-us_topic_0180094067_li1115151055511">Example 1: Create a training job parameter configuration using the data stored on OBS.<div class="codecoloring" codetype="Python" id="modelarts_04_0174__en-us_topic_0180094067_screen58470243552"><div class="highlight"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre><span class="normal"> 1</span>
<span class="normal"> 2</span>
<span class="normal"> 3</span>
<span class="normal"> 4</span>
<span class="normal"> 5</span>
<span class="normal"> 6</span>
<span class="normal"> 7</span>
<span class="normal"> 8</span>
<span class="normal"> 9</span>
<span class="normal">10</span>
<span class="normal">11</span>
<span class="normal">12</span>
<span class="normal">13</span>
<span class="normal">14</span>
<span class="normal">15</span>
<span class="normal">16</span>
<span class="normal">17</span>
<span class="normal">18</span>
<span class="normal">19</span>
<span class="normal">20</span></pre></div></td><td class="code"><div><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="kn">from</span> <span class="nn">modelarts.estimator</span> <span class="kn">import</span> <span class="n">Estimator</span>
<span class="n">session</span> <span class="o">=</span> <span class="n">Session</span><span class="p">()</span>
<span class="n">estimator</span> <span class="o">=</span> <span class="n">Estimator</span><span class="p">(</span>
<span class="n">modelarts_session</span><span class="o">=</span><span class="n">session</span><span class="p">,</span>
<span class="n">framework_type</span><span class="o">=</span><span class="s1">'PyTorch'</span><span class="p">,</span> <span class="c1"># AI engine name</span>
<span class="n">framework_version</span><span class="o">=</span><span class="s1">'PyTorch-1.0.0-python3.6'</span><span class="p">,</span> <span class="c1"># AI engine version</span>
<span class="n">code_dir</span><span class="o">=</span><span class="s1">'/bucket/src/'</span><span class="p">,</span> <span class="c1"># Training script directory</span>
<span class="n">boot_file</span><span class="o">=</span><span class="s1">'/bucket/src/pytorch_sentiment.py'</span><span class="p">,</span> <span class="c1"># Training boot script directory</span>
<span class="n">log_url</span><span class="o">=</span><span class="s1">'/bucket/log/'</span><span class="p">,</span> <span class="c1"># Training log directory</span>
<span class="n">hyperparameters</span><span class="o">=</span><span class="p">[</span>
<span class="p">{</span><span class="s2">&quot;label&quot;</span><span class="p">:</span><span class="s2">&quot;classes&quot;</span><span class="p">,</span>
<span class="s2">&quot;value&quot;</span><span class="p">:</span> <span class="s2">&quot;10&quot;</span><span class="p">},</span>
<span class="p">{</span><span class="s2">&quot;label&quot;</span><span class="p">:</span><span class="s2">&quot;lr&quot;</span><span class="p">,</span>
<span class="s2">&quot;value&quot;</span><span class="p">:</span> <span class="s2">&quot;0.001&quot;</span><span class="p">}</span>
<span class="p">],</span>
<span class="n">output_path</span><span class="o">=</span><span class="s1">'/bucket/output/'</span><span class="p">,</span> <span class="c1"># Training output directory</span>
<span class="n">train_instance_type</span><span class="o">=</span><span class="s1">'modelarts.vm.gpu.p100'</span><span class="p">,</span> <span class="c1"># Training environment flavor</span>
<span class="n">train_instance_count</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># Number of training nodes</span>
<span class="n">job_config_instance</span> <span class="o">=</span> <span class="n">estimator</span><span class="o">.</span><span class="n">create_job_configs</span><span class="p">(</span><span class="n">config_name</span><span class="o">=</span><span class="s1">'my_job_config'</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="s1">'/bucket/data/train/'</span><span class="p">,</span> <span class="n">config_desc</span><span class="o">=</span><span class="s1">'my job config'</span><span class="p">)</span>
</pre></div></td></tr></table></div>
</div>
</li></ul>
<ul id="modelarts_04_0174__en-us_topic_0180094067_ul15570131311556"><li id="modelarts_04_0174__en-us_topic_0180094067_li13570141313557">Example 2: Create a training job parameter configuration using a dataset.<div class="codecoloring" codetype="Python" id="modelarts_04_0174__en-us_topic_0180094067_screen74342036115513"><div class="highlight"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre><span class="normal"> 1</span>
<span class="normal"> 2</span>
<span class="normal"> 3</span>
<span class="normal"> 4</span>
<span class="normal"> 5</span>
<span class="normal"> 6</span>
<span class="normal"> 7</span>
<span class="normal"> 8</span>
<span class="normal"> 9</span>
<span class="normal">10</span>
<span class="normal">11</span>
<span class="normal">12</span>
<span class="normal">13</span>
<span class="normal">14</span>
<span class="normal">15</span>
<span class="normal">16</span>
<span class="normal">17</span>
<span class="normal">18</span>
<span class="normal">19</span>
<span class="normal">20</span></pre></div></td><td class="code"><div><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="kn">from</span> <span class="nn">modelarts.estimator</span> <span class="kn">import</span> <span class="n">Estimator</span>
<span class="n">session</span> <span class="o">=</span> <span class="n">Session</span><span class="p">()</span>
<span class="n">estimator</span> <span class="o">=</span> <span class="n">Estimator</span><span class="p">(</span>
<span class="n">modelarts_session</span><span class="o">=</span><span class="n">session</span><span class="p">,</span>
<span class="n">framework_type</span><span class="o">=</span><span class="s1">'PyTorch'</span><span class="p">,</span> <span class="c1"># AI engine name</span>
<span class="n">framework_version</span><span class="o">=</span><span class="s1">'PyTorch-1.0.0-python3.6'</span><span class="p">,</span> <span class="c1"># AI engine version</span>
<span class="n">code_dir</span><span class="o">=</span><span class="s1">'/bucket/src/'</span><span class="p">,</span> <span class="c1"># Training script directory</span>
<span class="n">boot_file</span><span class="o">=</span><span class="s1">'/bucket/src/pytorch_sentiment.py'</span><span class="p">,</span> <span class="c1"># Training boot script directory</span>
<span class="n">log_url</span><span class="o">=</span><span class="s1">'/bucket/log/'</span><span class="p">,</span> <span class="c1"># Training log directory</span>
<span class="n">hyperparameters</span><span class="o">=</span><span class="p">[</span>
<span class="p">{</span><span class="s2">&quot;label&quot;</span><span class="p">:</span><span class="s2">&quot;classes&quot;</span><span class="p">,</span>
<span class="s2">&quot;value&quot;</span><span class="p">:</span> <span class="s2">&quot;10&quot;</span><span class="p">},</span>
<span class="p">{</span><span class="s2">&quot;label&quot;</span><span class="p">:</span><span class="s2">&quot;lr&quot;</span><span class="p">,</span>
<span class="s2">&quot;value&quot;</span><span class="p">:</span> <span class="s2">&quot;0.001&quot;</span><span class="p">}</span>
<span class="p">],</span>
<span class="n">output_path</span><span class="o">=</span><span class="s1">'/bucket/output/'</span><span class="p">,</span> <span class="c1"># Training output directory</span>
<span class="n">train_instance_type</span><span class="o">=</span><span class="s1">'modelarts.vm.gpu.p100'</span><span class="p">,</span> <span class="c1"># Training environment flavor</span>
<span class="n">train_instance_count</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># Number of training nodes</span>
<span class="n">job_config_instance</span> <span class="o">=</span> <span class="n">estimator</span><span class="o">.</span><span class="n">create_job_configs</span><span class="p">(</span><span class="n">config_name</span><span class="o">=</span><span class="s1">'my_job_config'</span><span class="p">,</span> <span class="n">dataset_id</span><span class="o">=</span><span class="s1">'4AZNvFkN7KYr5EdhFkH'</span><span class="p">,</span> <span class="n">dataset_version_id</span><span class="o">=</span><span class="s1">'UOF9BIeSGArwVt0oI6T'</span><span class="p">,</span> <span class="n">config_desc</span><span class="o">=</span><span class="s1">'my job config'</span><span class="p">)</span>
</pre></div></td></tr></table></div>
</div>
</li></ul>
</div>
<div class="section" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_section0599140112517"><h4 class="sectiontitle">Parameter Description</h4>
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_table155461191218" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Estimator request parameters</caption><thead align="left"><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row254817912212"><th align="left" class="cellrowborder" valign="top" width="22.75%" id="mcps1.3.2.2.2.5.1.1"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p12549899214">Parameter</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="9.879999999999999%" id="mcps1.3.2.2.2.5.1.2"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p3552101193813">Mandatory</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="13.320000000000002%" id="mcps1.3.2.2.2.5.1.3"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1755169172118">Type</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="54.05%" id="mcps1.3.2.2.2.5.1.4"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p55521998211">Description</p>
</th>
</tr>
</thead>
<tbody><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row8893215413"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p6891421842">modelarts_session</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p68972047">Yes</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p158912219419">Object</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1689152543">Session object. For details about the initialization method, see <a href="modelarts_04_0123.html">Session Authentication</a>.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row197933582219"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p25545912114">train_instance_count</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p955361173817">Yes</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p55561095217">Long</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p25573982112">Number of workers in a training job</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row105532902114"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p115007506428">code_dir</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p4553121118384">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1750655034220">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p750985094216">Code directory of a training job, for example, <span class="filepath" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_filepath1951216503429"><b>/bucket/src/</b></span>. Leave this parameter blank when <strong id="modelarts_04_0174__en-us_topic_0180094067_b14311115102616">model_name</strong> is set.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row164861109396"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p35209501423">boot_file</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1955318112388">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p552245012421">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1052817508427">Boot file of a training job, which needs to be stored in the code directory. For example, <span class="filepath" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_filepath953055064213"><b>/bucket/src/boot.py</b></span>. Leave this parameter blank when <strong id="modelarts_04_0174__en-us_topic_0180094067_b31141431172612">model_name</strong> is set.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row262591093918"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1655814915214">model_name</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1155391163819">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p55598942112">Long</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p115621942111">Name of the built-in algorithm used by a training job. If you have configured <strong id="modelarts_04_0174__en-us_topic_0180094067_b145121039182618">model_name</strong>, you do not need to configure <strong id="modelarts_04_0174__en-us_topic_0180094067_b1512339142610">app_url</strong>, <strong id="modelarts_04_0174__en-us_topic_0180094067_b1451319392262">boot_file_url</strong>, <strong id="modelarts_04_0174__en-us_topic_0180094067_b6513103982612">framework_type</strong>, and <strong id="modelarts_04_0174__en-us_topic_0180094067_b11514939202617">framework_version</strong>.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row1390105335719"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p7801836172112">output_path</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p95531511163817">Yes</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p12801133632113">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p880183622110">Output path of a training job</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row3799183612114"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p13562119132119">hyperparameters</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p655755054219">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1756418919215">JSON Array</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1956514920216">Running parameters of a training job. It is a collection of label-value pairs. This parameter is a container environment variable when a job uses a custom image.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row83521247152016"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p106881050154220">log_url</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p17806757142413">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1369555013425">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p9699950194211">OBS URL of the logs of a training job. By default, this parameter is left blank. Example value: <span class="filepath" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_filepath770045019429"><b>/usr/log/</b></span></p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row1912536142111"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p109129363215">train_instance_type</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p555331113819">Yes</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p291283612219">Long</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p199131536192118">Resource flavor selected for a training job. If you choose to train on the training platform, obtain the value by calling the API described in <a href="modelarts_04_0191.html">Querying the List of Resource Flavors</a>.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row13142103743519"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p2014213371353">framework_type</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p16553201143810">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1014217373359">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p565095012422">Engine selected for a training job. Obtain the value by calling the API described in <a href="modelarts_04_0192.html">Querying the List of Engine Types</a>. Leave this parameter blank when <strong id="modelarts_04_0174__en-us_topic_0180094067_b127142715272">model_name</strong> is set.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row192212012216"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1822211112118">framework_version</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1722201162119">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p142221612219">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p6270941135920">Engine version selected for a training job. Obtain the value by calling the API described in <a href="modelarts_04_0192.html">Querying the List of Engine Types</a>. Leave this parameter blank when <strong id="modelarts_04_0174__en-us_topic_0180094067_b13363714172817">model_name</strong> is set.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row9499151142612"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p12675171136">job_description</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p196751771039">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p8675972311">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1767517712315">Description of a training job</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row49082484413"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p255514164311">user_image_url</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p5555144113435">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p5555841114318">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p755554118434">SWR URL of the custom image used by a training job. Example value: <span class="filepath" id="modelarts_04_0174__en-us_topic_0180094067_filepath45311015165614"><b>100.125.5.235:20202/jobmng/custom-cpu-base:1.0</b></span></p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row19129155110419"><td class="cellrowborder" valign="top" width="22.75%" headers="mcps1.3.2.2.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p46411034174620">user_command</p>
</td>
<td class="cellrowborder" valign="top" width="9.879999999999999%" headers="mcps1.3.2.2.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p11641434154610">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.320000000000002%" headers="mcps1.3.2.2.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1364113434619">String</p>
</td>
<td class="cellrowborder" valign="top" width="54.05%" headers="mcps1.3.2.2.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p17641113410465">Boot command used to start the container of the custom image of a training job. The format is <span class="filepath" id="modelarts_04_0174__en-us_topic_0180094067_filepath12983142420569"><b>bash /home/work/run_train.sh python /home/work/user-job-dir/app/train.py {python_file_parameter}</b></span>.</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_table1218411161451" frame="border" border="1" rules="all"><caption><b>Table 2 </b><strong id="modelarts_04_0174__en-us_topic_0180094067_b732412341280">create_job_configs</strong> request parameters</caption><thead align="left"><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row1184151614517"><th align="left" class="cellrowborder" valign="top" width="23.232323232323232%" id="mcps1.3.3.2.5.1.1"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p118471620514">Parameter</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="9.36093609360936%" id="mcps1.3.3.2.5.1.2"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p181843161754">Mandatory</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="13.831383138313832%" id="mcps1.3.3.2.5.1.3"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p12184316451">Type</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="53.57535753575357%" id="mcps1.3.3.2.5.1.4"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p618413161353">Description</p>
</th>
</tr>
</thead>
<tbody><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row2018551610513"><td class="cellrowborder" valign="top" width="23.232323232323232%" headers="mcps1.3.3.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p51851016153">config_name</p>
</td>
<td class="cellrowborder" valign="top" width="9.36093609360936%" headers="mcps1.3.3.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p12185716856">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.831383138313832%" headers="mcps1.3.3.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p518561618518">String</p>
</td>
<td class="cellrowborder" valign="top" width="53.57535753575357%" headers="mcps1.3.3.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p8185216258">Name of a training job parameter configuration. The value is a string of 1 to 20 characters consisting of only digits, letters, underscores (_), and hyphens (-). By default, if this parameter is left blank, the value is dynamically generated by date.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row1718511161058"><td class="cellrowborder" valign="top" width="23.232323232323232%" headers="mcps1.3.3.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p51853164515">config_desc</p>
</td>
<td class="cellrowborder" valign="top" width="9.36093609360936%" headers="mcps1.3.3.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1518519161057">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.831383138313832%" headers="mcps1.3.3.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p81857161653">String</p>
</td>
<td class="cellrowborder" valign="top" width="53.57535753575357%" headers="mcps1.3.3.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p151859164510">Description of a training job parameter configuration. The value is a string of 0 to 256 characters. By default, this parameter is left blank.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row10185916358"><td class="cellrowborder" valign="top" width="23.232323232323232%" headers="mcps1.3.3.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p018541619520">inputs</p>
</td>
<td class="cellrowborder" valign="top" width="9.36093609360936%" headers="mcps1.3.3.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p101856164512">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.831383138313832%" headers="mcps1.3.3.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p718512161854">String</p>
</td>
<td class="cellrowborder" valign="top" width="53.57535753575357%" headers="mcps1.3.3.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p18185816458">OBS storage path of a training job</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row8806165254911"><td class="cellrowborder" valign="top" width="23.232323232323232%" headers="mcps1.3.3.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1679405218495">dataset_id</p>
</td>
<td class="cellrowborder" valign="top" width="9.36093609360936%" headers="mcps1.3.3.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p5794105224916">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.831383138313832%" headers="mcps1.3.3.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p27946521492">String</p>
</td>
<td class="cellrowborder" valign="top" width="53.57535753575357%" headers="mcps1.3.3.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p07941452174917">Dataset ID of a training job. This parameter must be used together with <strong id="modelarts_04_0174__en-us_topic_0180094067_b1382904062410">dataset_version_id</strong>, but cannot be used together with <strong id="modelarts_04_0174__en-us_topic_0180094067_b2832440172419">inputs</strong>.</p>
</td>
</tr>
<tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_row138061252144911"><td class="cellrowborder" valign="top" width="23.232323232323232%" headers="mcps1.3.3.2.5.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p179475218491">dataset_version_id</p>
</td>
<td class="cellrowborder" valign="top" width="9.36093609360936%" headers="mcps1.3.3.2.5.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p779415284913">No</p>
</td>
<td class="cellrowborder" valign="top" width="13.831383138313832%" headers="mcps1.3.3.2.5.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1279415522491">String</p>
</td>
<td class="cellrowborder" valign="top" width="53.57535753575357%" headers="mcps1.3.3.2.5.1.4 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0170904401_p1779445294919">Dataset version ID of a training job. This parameter must be used together with <strong id="modelarts_04_0174__en-us_topic_0180094067_b1674324172910">dataset_id</strong>, but cannot be used together with <strong id="modelarts_04_0174__en-us_topic_0180094067_b16758240293">inputs</strong>.</p>
</td>
</tr>
</tbody>
</table>
</div>
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_table973120224596" frame="border" border="1" rules="all"><caption><b>Table 3 </b><strong id="modelarts_04_0174__en-us_topic_0180094067_b447982514291">create_job_configs</strong> response parameters</caption><thead align="left"><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_row2731522195910"><th align="left" class="cellrowborder" valign="top" width="24.91249124912491%" id="mcps1.3.4.2.4.1.1"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p57306225598">Parameter</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="21.732173217321733%" id="mcps1.3.4.2.4.1.2"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p117308225593">Type</p>
</th>
<th align="left" class="cellrowborder" valign="top" width="53.35533553355335%" id="mcps1.3.4.2.4.1.3"><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p2730132255915">Description</p>
</th>
</tr>
</thead>
<tbody><tr id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_row1873172215912"><td class="cellrowborder" valign="top" width="24.91249124912491%" headers="mcps1.3.4.2.4.1.1 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p4731322145919">TrainingJob</p>
</td>
<td class="cellrowborder" valign="top" width="21.732173217321733%" headers="mcps1.3.4.2.4.1.2 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p117311922115916">Object</p>
</td>
<td class="cellrowborder" valign="top" width="53.35533553355335%" headers="mcps1.3.4.2.4.1.3 "><p id="modelarts_04_0174__en-us_topic_0180094067_en-us_topic_0160436006_p6731182225914">Training object. This object contains attributes such as <strong id="modelarts_04_0174__en-us_topic_0180094067_b2712185914289">config_name</strong>, and operations on a training job parameter configuration, such as querying or deleting the training job parameter configuration. For example, you can use <strong id="modelarts_04_0174__en-us_topic_0180094067_b87138592288">job_config_instance.config_name</strong> to obtain the name of a training job parameter configuration.</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
<div>
<div class="familylinks">
<div class="parentlink"><strong>Parent topic:</strong> <a href="modelarts_04_0173.html">Training Job Parameter Configuration</a></div>
</div>
</div>
View Git Blame Copy Permalink