http://git-wip-us.apache.org/repos/asf/predictionio-site/blob/c17b9607/templates/productranking/quickstart/index.html ---------------------------------------------------------------------- diff --git a/templates/productranking/quickstart/index.html b/templates/productranking/quickstart/index.html index 8c2bb6f..3e9ee50 100644 --- a/templates/productranking/quickstart/index.html +++ b/templates/productranking/quickstart/index.html @@ -32,7 +32,7 @@ Your system is all ready to go. <span class="o">[</span>INFO] <span class="o">[</span>App<span class="nv">$]</span> MyApp1 | 1 | 3mZWDzci2D5YsqAnqNnXH9SB6Rg3dsTBs8iHkK6X2i54IQsIZI1eEeQQyMfs7b3F | <span class="o">(</span>all<span class="o">)</span> <span class="o">[</span>INFO] <span class="o">[</span>App<span class="nv">$]</span> MyApp2 | 2 | io5lz6Eg4m3Xe4JZTBFE13GMAf1dhFl6ZteuJfrO84XpdOz9wRCrDU44EUaYuXq5 | <span class="o">(</span>all<span class="o">)</span> <span class="o">[</span>INFO] <span class="o">[</span>App<span class="nv">$]</span> Finished listing 2 app<span class="o">(</span>s<span class="o">)</span>. -</pre></td></tr></tbody></table> </div> <p><a href="#"></a></p> <h2 id='4.-collecting-data' class='header-anchors'>4. Collecting Data</h2><p>Next, let's collect some training data for the app of this Engine. By default, the Product Ranking Engine Template supports 2 types of entities: <strong>user</strong> and <strong>item</strong>, and event <strong>view</strong>. A user can view an item. This template requires '$set' user event, '$set' item event and user-view-item events.</p> <p>You can send these events to PredictionIO Event Server in real-time easily by making a HTTP request or through the provided SDK. Please see <a href="/appintegration/">App Integration Overview</a> for more details how to integrate your app with SDK.</p><p>Let's try sending events to EventServer with the following <code>curl</code> commands (The corresponding SDK code is showed in other tabs).</p><p>Replace <code><ACCCESS_KEY></code> by the Access Key generated in above steps. Note that <code>localhost:7070</code> is the default URL of the Event Server.</p><p>For convenience, set your access key to the shell variable, run:</p><p><code>$ ACCESS_KEY=<ACCESS_KEY></code></p> <p><a href="#"></a></p> <p>For example, when a new user with id "u0" is created in your app on time <code>2014-11-02T09:39:45.618-08:00</code> (current time will be used if eventTime is not specified), you can send a <code>$set</code> event for this user. To send this event, run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-f6c7ce95-0cf0-45b3-a2fb-ae39a3f662ac">REST API</a></li> <li data-lang="python"><a href="#tab-316df7bc-9424-4830-ba9f-c6d4aee0cafd">Python SDK</a></li> <li data-lang="php"><a href="#tab-b44618aa-c555-43cb-bfda-016e49ffdbe2">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-8764afe4-ca50-4c6b-8ceb-4519fc045f35">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-c9f5bff6-c898-45f6 -ab1f-0ec3fd6ca056">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-f6c7ce95-0cf0-45b3-a2fb-ae39a3f662ac"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td></tr></tbody></table> </div> <p><a href="#"></a></p> <h2 id='4.-collecting-data' class='header-anchors'>4. Collecting Data</h2><p>Next, let's collect some training data for the app of this Engine. By default, the Product Ranking Engine Template supports 2 types of entities: <strong>user</strong> and <strong>item</strong>, and event <strong>view</strong>. A user can view an item. This template requires '$set' user event, '$set' item event and user-view-item events.</p> <p>You can send these events to PredictionIO Event Server in real-time easily by making a HTTP request or through the provided SDK. Please see <a href="/appintegration/">App Integration Overview</a> for more details how to integrate your app with SDK.</p><p>Let's try sending events to EventServer with the following <code>curl</code> commands (The corresponding SDK code is showed in other tabs).</p><p>Replace <code><ACCCESS_KEY></code> by the Access Key generated in above steps. Note that <code>localhost:7070</code> is the default URL of the Event Server.</p><p>For convenience, set your access key to the shell variable, run:</p><p><code>$ ACCESS_KEY=<ACCESS_KEY></code></p> <p><a href="#"></a></p> <p>For example, when a new user with id "u0" is created in your app on time <code>2014-11-02T09:39:45.618-08:00</code> (current time will be used if eventTime is not specified), you can send a <code>$set</code> event for this user. To send this event, run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-a870a4b2-9b66-4d27-ba2f-b0b5b77a9b80">REST API</a></li> <li data-lang="python"><a href="#tab-465b7cf9-5be4-4634-b934-5bbbe0df0a84">Python SDK</a></li> <li data-lang="php"><a href="#tab-804a0f50-54f6-4cfb-8825-aec74c5517c7">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-6c5fee9e-50a1-4e82-8cec-5c06be2d7aaa">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-61197684-19e5-4206 -ae56-3792784c5600">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-a870a4b2-9b66-4d27-ba2f-b0b5b77a9b80"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -47,7 +47,7 @@ Your system is all ready to go. "entityId" : "u0", "eventTime" : "2014-11-02T09:39:45.618-08:00" }'</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-316df7bc-9424-4830-ba9f-c6d4aee0cafd"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-465b7cf9-5be4-4634-b934-5bbbe0df0a84"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -78,7 +78,7 @@ Your system is all ready to go. <span class="n">entity_type</span><span class="o">=</span><span class="s">"user"</span><span class="p">,</span> <span class="n">entity_id</span><span class="o">=<</span><span class="n">USER_ID</span><span class="o">></span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-b44618aa-c555-43cb-bfda-016e49ffdbe2"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-804a0f50-54f6-4cfb-8825-aec74c5517c7"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -103,7 +103,7 @@ Your system is all ready to go. <span class="s1">'entityId'</span> <span class="o">=></span> <span class="o"><</span><span class="nx">USER</span> <span class="nx">ID</span><span class="o">></span> <span class="p">));</span> <span class="cp">?></span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-8764afe4-ca50-4c6b-8ceb-4519fc045f35"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-6c5fee9e-50a1-4e82-8cec-5c06be2d7aaa"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -120,7 +120,7 @@ Your system is all ready to go. <span class="s1">'user'</span><span class="p">,</span> <span class="o"><</span><span class="no">USER</span> <span class="no">ID</span><span class="o">></span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-c9f5bff6-c898-45f6-ab1f-0ec3fd6ca056"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-61197684-19e5-4206-ae56-3792784c5600"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -145,7 +145,7 @@ Your system is all ready to go. <span class="o">.</span><span class="na">entityType</span><span class="o">(</span><span class="s">"user"</span><span class="o">)</span> <span class="o">.</span><span class="na">entityId</span><span class="o">(<</span><span class="n">USER_ID</span><span class="o">>);</span> <span class="n">client</span><span class="o">.</span><span class="na">createEvent</span><span class="o">(</span><span class="n">userEvent</span><span class="o">);</span> -</pre></td> </tr></tbody></table> </div> </div> </div> <p>When a new item "i0" is created in your app on time <code>2014-11-02T09:39:45.618-08:00</code> (current time will be used if eventTime is not specified), you can send a <code>$set</code> event for the item. Run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-c123e2b4-ec0c-4747-afe3-99970dd8d457">REST API</a></li> <li data-lang="python"><a href="#tab-687f83e3-f39a-44fa-a360-5a7e346f7df2">Python SDK</a></li> <li data-lang="php"><a href="#tab-9b08ae83-4061-4a67-a6ba-583868b277f9">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-234fbf42-8a23-488a-8a6a-154a498a2bf9">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-534577d4-b2be-41b1-8e9d-a7088ed5929f">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-c123e2b4-ec0c-4747-afe3-99970dd8d457"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td cla ss="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> </div> <p>When a new item "i0" is created in your app on time <code>2014-11-02T09:39:45.618-08:00</code> (current time will be used if eventTime is not specified), you can send a <code>$set</code> event for the item. Run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-edc4d81a-2db0-4a4d-a5fe-1c58088624b2">REST API</a></li> <li data-lang="python"><a href="#tab-6f619515-acdd-4ee4-88de-8789b828d954">Python SDK</a></li> <li data-lang="php"><a href="#tab-c91ba414-f24f-4677-a461-41e17b474c7a">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-316c0bec-2f4c-49f9-9c8a-b714828ddc5e">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-4fb970eb-67d9-4939-8700-c1ecbdc9357b">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-edc4d81a-2db0-4a4d-a5fe-1c58088624b2"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td cla ss="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -160,7 +160,7 @@ Your system is all ready to go. "entityId" : "i0", "eventTime" : "2014-11-02T09:39:45.618-08:00" }'</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-687f83e3-f39a-44fa-a360-5a7e346f7df2"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-6f619515-acdd-4ee4-88de-8789b828d954"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -173,7 +173,7 @@ Your system is all ready to go. <span class="n">entity_type</span><span class="o">=</span><span class="s">"item"</span><span class="p">,</span> <span class="n">entity_id</span><span class="o">=<</span><span class="n">ITEM_ID</span><span class="o">></span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-9b08ae83-4061-4a67-a6ba-583868b277f9"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-c91ba414-f24f-4677-a461-41e17b474c7a"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -190,7 +190,7 @@ Your system is all ready to go. <span class="p">));</span> <span class="cp">?></span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-234fbf42-8a23-488a-8a6a-154a498a2bf9"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-316c0bec-2f4c-49f9-9c8a-b714828ddc5e"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -203,7 +203,7 @@ Your system is all ready to go. <span class="o"><</span><span class="no">ITEM</span> <span class="no">ID</span><span class="o">></span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-534577d4-b2be-41b1-8e9d-a7088ed5929f"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-4fb970eb-67d9-4939-8700-c1ecbdc9357b"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -214,7 +214,7 @@ Your system is all ready to go. <span class="o">.</span><span class="na">entityType</span><span class="o">(</span><span class="s">"item"</span><span class="o">)</span> <span class="o">.</span><span class="na">entityId</span><span class="o">(<</span><span class="n">ITEM_ID</span><span class="o">>);</span> <span class="n">client</span><span class="o">.</span><span class="na">createEvent</span><span class="o">(</span><span class="n">itemEvent</span><span class="o">)</span> -</pre></td> </tr></tbody></table> </div> </div> </div> <p>When the user "u0" view item "i0" on time <code>2014-11-10T12:34:56.123-08:00</code> (current time will be used if eventTime is not specified), you can send a view event. Run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-c2ff15f5-714e-4235-9603-e912637e0ca4">REST API</a></li> <li data-lang="python"><a href="#tab-28a67892-4244-4124-9964-271897eabc92">Python SDK</a></li> <li data-lang="php"><a href="#tab-5c824856-ab08-47c2-9099-7713f336caaf">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-20bfad51-2935-47a2-bb14-63e6d6aac973">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-c08337e0-539f-4567-8121-eaf956e42bc3">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-c2ff15f5-714e-4235-9603-e912637e0ca4"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text -align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> </div> <p>When the user "u0" view item "i0" on time <code>2014-11-10T12:34:56.123-08:00</code> (current time will be used if eventTime is not specified), you can send a view event. Run the following <code>curl</code> command:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-b2f6de76-4058-41e9-b6db-30d5b900f25f">REST API</a></li> <li data-lang="python"><a href="#tab-139a990a-3269-4a72-a2a7-9c83dab0658d">Python SDK</a></li> <li data-lang="php"><a href="#tab-143a1e87-3443-4e4a-9a16-4041886f4cd0">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-7faa37dc-3aaf-44fb-9677-0db6cc0a9808">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-ce687ae9-f404-4464-9c7b-fa07cb439159">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-b2f6de76-4058-41e9-b6db-30d5b900f25f"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text -align: right"><pre class="lineno">1 2 3 4 @@ -233,7 +233,7 @@ Your system is all ready to go. "targetEntityId" : "i0", "eventTime" : "2014-11-10T12:34:56.123-08:00" }'</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-28a67892-4244-4124-9964-271897eabc92"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-139a990a-3269-4a72-a2a7-9c83dab0658d"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -250,7 +250,7 @@ Your system is all ready to go. <span class="n">target_entity_type</span><span class="o">=</span><span class="s">"item"</span><span class="p">,</span> <span class="n">target_entity_id</span><span class="o">=<</span><span class="n">ITEM</span> <span class="n">ID</span><span class="o">></span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-5c824856-ab08-47c2-9099-7713f336caaf"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-143a1e87-3443-4e4a-9a16-4041886f4cd0"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -269,7 +269,7 @@ Your system is all ready to go. <span class="s1">'targetEntityId'</span> <span class="o">=></span> <span class="o"><</span><span class="nx">ITEM</span> <span class="nx">ID</span><span class="o">></span> <span class="p">));</span> <span class="cp">?></span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-20bfad51-2935-47a2-bb14-63e6d6aac973"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-7faa37dc-3aaf-44fb-9677-0db6cc0a9808"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -286,7 +286,7 @@ Your system is all ready to go. <span class="s1">'targetEntityId'</span> <span class="o">=></span> <span class="o"><</span><span class="no">ITEM</span> <span class="no">ID</span><span class="o">></span> <span class="p">}</span> <span class="p">)</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-c08337e0-539f-4567-8121-eaf956e42bc3"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-ce687ae9-f404-4464-9c7b-fa07cb439159"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -344,14 +344,14 @@ User u10 views item i7 </pre></td></tr></tbody></table> </div> <p>When the engine is deployed successfully and running, you should see a console message similar to the following:</p><div class="highlight shell"><table style="border-spacing: 0"><tbody><tr><td class="gutter gl" style="text-align: right"><pre class="lineno">1 2</pre></td><td class="code"><pre><span class="o">[</span>INFO] <span class="o">[</span>HttpListener] Bound to /0.0.0.0:8000 <span class="o">[</span>INFO] <span class="o">[</span>MasterActor] Bind successful. Ready to serve. -</pre></td></tr></tbody></table> </div> <p>Do not kill the deployed engine process.</p><p>By default, the deployed engine binds to <a href="http://localhost:8000";><a href="http://localhost:8000";>http://localhost:8000</a></a>. You can visit that page in your web browser to check its status.</p><p><img alt="Engine Status" src="/images/engine-server-3246414b.png"/></p></p><h2 id='6.-use-the-engine' class='header-anchors'>6. Use the Engine</h2><p>Now, You can retrieve predicted results. To rank items with IDs "i1", "i3", "i10", "i2", "i5", "i31", "i9" for user "u2". You send this JSON <code>{ "user": "u2", "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"] }</code> to the deployed engine and it will return a JSON of the ranked items. Simply send a query by making a HTTP request or through the <code>EngineCl ient</code> of an SDK.</p><p>With the deployed engine running, open another terminal and run the following <code>curl</code> command or use SDK to send the query:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-e21bb54d-c52b-47d8-972f-1d5b92593ec0">REST API</a></li> <li data-lang="python"><a href="#tab-92d50e54-9aaa-4e72-8a18-d667572ac2e8">Python SDK</a></li> <li data-lang="php"><a href="#tab-7c13ef7b-8216-4f7d-b03d-65a1c5f2665b">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-e92ec963-5243-4026-a73c-fef3821dd823">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-c6ae3acd-18c6-4285-a4e6-773feec019c0">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-e21bb54d-c52b-47d8-972f-1d5b92593ec0"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td></tr></tbody></table> </div> <p>Do not kill the deployed engine process.</p><p>By default, the deployed engine binds to <a href="http://localhost:8000";><a href="http://localhost:8000";>http://localhost:8000</a></a>. You can visit that page in your web browser to check its status.</p><p><img alt="Engine Status" src="/images/engine-server-3246414b.png"/></p></p><h2 id='6.-use-the-engine' class='header-anchors'>6. Use the Engine</h2><p>Now, You can retrieve predicted results. To rank items with IDs "i1", "i3", "i10", "i2", "i5", "i31", "i9" for user "u2". You send this JSON <code>{ "user": "u2", "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"] }</code> to the deployed engine and it will return a JSON of the ranked items. Simply send a query by making a HTTP request or through the <code>EngineCl ient</code> of an SDK.</p><p>With the deployed engine running, open another terminal and run the following <code>curl</code> command or use SDK to send the query:</p><div class="tabs"> <ul class="control"> <li data-lang="json"><a href="#tab-2426c8a7-b0bd-4ec1-9e9d-93fadca970a6">REST API</a></li> <li data-lang="python"><a href="#tab-ae7e0c2d-ffcd-480f-bbd9-9a92a87c9d1b">Python SDK</a></li> <li data-lang="php"><a href="#tab-7600643d-c044-40b8-af28-3e1223f41d24">PHP SDK</a></li> <li data-lang="ruby"><a href="#tab-e26dbd1d-1509-4aad-95a6-369834b62185">Ruby SDK</a></li> <li data-lang="java"><a href="#tab-e6eb9cef-be5d-4270-9834-3536af7f2e47">Java SDK</a></li> </ul> <div data-tab="REST API" data-lang="json" id="tab-2426c8a7-b0bd-4ec1-9e9d-93fadca970a6"> <div class="highlight shell"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4</pre></td> <td class="code"><pre><span class="gp">$ </span>curl -H <span class="s2">"Content-Type: application/json"</span> <span class="se">\</span> -d <span class="s1">'{ "user": "u2", "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"]}'</span> <span class="se">\</span> http://localhost:8000/queries.json -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-92d50e54-9aaa-4e72-8a18-d667572ac2e8"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Python SDK" data-lang="python" id="tab-ae7e0c2d-ffcd-480f-bbd9-9a92a87c9d1b"> <div class="highlight python"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -362,7 +362,7 @@ http://localhost:8000/queries.json <span class="s">"user"</span><span class="p">:</span> <span class="s">"u2"</span><span class="p">,</span> <span class="s">"items"</span><span class="p">:</span> <span class="p">[</span><span class="s">"i1"</span><span class="p">,</span> <span class="s">"i3"</span><span class="p">,</span> <span class="s">"i10"</span><span class="p">,</span> <span class="s">"i2"</span><span class="p">,</span> <span class="s">"i5"</span><span class="p">,</span> <span class="s">"i31"</span><span class="p">,</span> <span class="s">"i9"</span><span class="p">]</span> <span class="p">})</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-7c13ef7b-8216-4f7d-b03d-65a1c5f2665b"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="PHP SDK" data-lang="php" id="tab-7600643d-c044-40b8-af28-3e1223f41d24"> <div class="highlight php"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -385,7 +385,7 @@ http://localhost:8000/queries.json <span class="nb">print_r</span><span class="p">(</span><span class="nv">$response</span><span class="p">);</span> <span class="cp">?></span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-e92ec963-5243-4026-a73c-fef3821dd823"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Ruby SDK" data-lang="ruby" id="tab-e26dbd1d-1509-4aad-95a6-369834b62185"> <div class="highlight ruby"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 @@ -402,7 +402,7 @@ http://localhost:8000/queries.json <span class="s1">'items'</span> <span class="o">=></span> <span class="p">[</span><span class="s1">'i1'</span><span class="p">,</span> <span class="s1">'i3'</span><span class="p">,</span> <span class="s1">'i10'</span><span class="p">,</span> <span class="s1">'i2'</span><span class="p">,</span> <span class="s1">'i5'</span><span class="p">,</span> <span class="s1">'i31'</span><span class="p">,</span> <span class="s1">'i9'</span><span class="p">])</span> <span class="nb">puts</span> <span class="n">response</span> -</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-c6ae3acd-18c6-4285-a4e6-773feec019c0"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td> </tr></tbody></table> </div> </div> <div data-tab="Java SDK" data-lang="java" id="tab-e6eb9cef-be5d-4270-9834-3536af7f2e47"> <div class="highlight java"> <table style="border-spacing: 0"><tbody><tr> <td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4
http://git-wip-us.apache.org/repos/asf/predictionio-site/blob/c17b9607/templates/recommendation/batch-evaluator/index.html ---------------------------------------------------------------------- diff --git a/templates/recommendation/batch-evaluator/index.html b/templates/recommendation/batch-evaluator/index.html index 392eb7a..d6888ca 100644 --- a/templates/recommendation/batch-evaluator/index.html +++ b/templates/recommendation/batch-evaluator/index.html @@ -59,7 +59,7 @@ <span class="nc">Seq</span><span class="o">(</span><span class="n">evalDataSet</span><span class="o">)</span> <span class="o">}</span> -</pre></td></tr></tbody></table> </div> <div class="alert-message note"><p>Alternatively, you can create a new DataSource extending original DataSource. Then you can add the new one in Engine.scala and specify which one to use in engine.json.</p></div> <h2 id='2.-add-a-new-evaluator' class='header-anchors'>2. Add a new Evaluator</h2><p>Create a new file <code>BatchPersistableEvaluator.scala</code>. Unlike the <code>MetricEvaluator</code>, this Evaluator simply writes the Query and correpsonding PredictedResult to the output directory without performaning any metrics calculation.</p><p>Note that output directory is specified by the variable <code>outputDir</code>.</p><div class="highlight scala"><table style="border-spacing: 0"><tbody><tr><td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td></tr></tbody></table> </div> <div class="alert-message note"><p>Alternatively, you can create a new DataSource extending original DataSource. Then you can add the new one in Engine.scala and specify which one to use in engine.json.</p></div> <h2 id='2.-add-a-new-evaluator' class='header-anchors'>2. Add a new Evaluator</h2><p>Create a new file <code>BatchPersistableEvaluator.scala</code>. Unlike the <code>MetricEvaluator</code>, this Evaluator simply writes the Query and corresponding PredictedResult to the output directory without performing any metrics calculation.</p><p>Note that output directory is specified by the variable <code>outputDir</code>.</p><div class="highlight scala"><table style="border-spacing: 0"><tbody><tr><td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3 4 http://git-wip-us.apache.org/repos/asf/predictionio-site/blob/c17b9607/templates/recommendation/evaluation/index.html ---------------------------------------------------------------------- diff --git a/templates/recommendation/evaluation/index.html b/templates/recommendation/evaluation/index.html index 933407a..3cd1dc2 100644 --- a/templates/recommendation/evaluation/index.html +++ b/templates/recommendation/evaluation/index.html @@ -111,7 +111,7 @@ Metrics: PositiveCount <span class="o">(</span><span class="nv">threshold</span><span class="o">=</span>1.0<span class="o">)</span>: 10.006666666666666 <span class="o">[</span>INFO 2015-03-31 00:36:01,516] <span class="o">[</span>CoreWorkflow<span class="nv">$]</span> runEvaluation completed -</pre></td></tr></tbody></table> </div> <p>The console prints out the evaluation meric score of each engine params, and finally pretty print the optimal engine params. Amongs the 3 engine params we evaluate, the best Prediction@k has a score of ~0.1521.</p><h2 id='the-evaluation-design' class='header-anchors'>The Evaluation Design</h2><p>We assume you have read the <a href="/evaluation">Tuning and Evaluation</a> section. We will cover the evaluation aspects which are specific to the recommendation engine.</p><p>In recommendation evaluation, the raw data is a sequence of known ratings. A rating has 3 components: user, item, and a score. We use the $k-fold$ method for evaluation, the raw data is sliced into a sequence of (training, validation) data tuple.</p><p>In the validation data, we construct a query for <em>each user</em>, and get a list of recommended items from the engine. It is vastly different from the classification tutorial, where there is a one-to-one corresponding betwee n the training data point and the validation data point. In this evaluation, our unit of evaluation is <em>user</em>, we evaluate the quality of an engine using the known rating of a user.</p><h3 id='key-assumptions' class='header-anchors'>Key assumptions</h3><p>There are multiple assumptions we have to make when we evaluate a recommendation engine:</p> <ul> <li><p>Definition of 'good'. We want to quantify if the engine is able to recommend items which the user likes, we need to define what is meant by 'good'. In this examle, we have two kinds of events: 'rate' and 'buy'. The 'rate' event is associated with a rating value which ranges between 1 to 4, and the 'buy' event is mapped to a rating of 4. When we implement the metric, we have to specify a rating threshold, only the rating above the threshold is considered 'good'.</p></li> <li><p>The absense of complete rating. It is extremely unlikely that the training data contains ra ting for all user-item tuples. In contrast, of a system containing 1000 items, a user may only have rated 20 of them, leaving 980 items unrated. There is no way for us to certainly tell if the user likes an unrated product. When we examine the evaluation result, it is important for us to keep in mind that the final metric is only an approximation of the actual result.</p></li> <li><p>Recommendation affects user behavior. Suppose you are a e-commerce company and would like to use the recommendation engine to personalize the landing page, the item you show in the langing page directly impacts what the user is going to purchase. This is different from weather prediction, whatever the weather forecast engine predicts, tomorrow's weather won't be affected. Therefore, when we conduct offline evaluation for recommendation engines, it is possible that the final user behavior is dramatically different from the evaluation result. However, in the evaluation, for simplicity, we have to assume that user behavior is homogenous.</p></li> </ul> <h2 id='evaluation-data-generation' class='header-anchors'>Evaluation Data Generation</h2><h3 id='actual-result' class='header-anchors'>Actual Result</h3><p>In MyRecommendation/src/main/scala/<strong><em>Engine.scala</em></strong>, we define the <code>ActualResult</code> which represents the user rating for validation. It stores the list of ratings in the validation set for a user.</p><div class="highlight scala"><table style="border-spacing: 0"><tbody><tr><td class="gutter gl" style="text-align: right"><pre class="lineno">1 +</pre></td></tr></tbody></table> </div> <p>The console prints out the evaluation meric score of each engine params, and finally pretty print the optimal engine params. Amongs the 3 engine params we evaluate, the best Prediction@k has a score of ~0.1521.</p><h2 id='the-evaluation-design' class='header-anchors'>The Evaluation Design</h2><p>We assume you have read the <a href="/evaluation">Tuning and Evaluation</a> section. We will cover the evaluation aspects which are specific to the recommendation engine.</p><p>In recommendation evaluation, the raw data is a sequence of known ratings. A rating has 3 components: user, item, and a score. We use the $k-fold$ method for evaluation, the raw data is sliced into a sequence of (training, validation) data tuple.</p><p>In the validation data, we construct a query for <em>each user</em>, and get a list of recommended items from the engine. It is vastly different from the classification tutorial, where there is a one-to-one corresponding betwee n the training data point and the validation data point. In this evaluation, our unit of evaluation is <em>user</em>, we evaluate the quality of an engine using the known rating of a user.</p><h3 id='key-assumptions' class='header-anchors'>Key assumptions</h3><p>There are multiple assumptions we have to make when we evaluate a recommendation engine:</p> <ul> <li><p>Definition of 'good'. We want to quantify if the engine is able to recommend items which the user likes, we need to define what is meant by 'good'. In this examle, we have two kinds of events: 'rate' and 'buy'. The 'rate' event is associated with a rating value which ranges between 1 to 4, and the 'buy' event is mapped to a rating of 4. When we implement the metric, we have to specify a rating threshold, only the rating above the threshold is considered 'good'.</p></li> <li><p>The absence of complete rating. It is extremely unlikely that the training data contains ra ting for all user-item tuples. In contrast, of a system containing 1000 items, a user may only have rated 20 of them, leaving 980 items unrated. There is no way for us to certainly tell if the user likes an unrated product. When we examine the evaluation result, it is important for us to keep in mind that the final metric is only an approximation of the actual result.</p></li> <li><p>Recommendation affects user behavior. Suppose you are a e-commerce company and would like to use the recommendation engine to personalize the landing page, the item you show in the langing page directly impacts what the user is going to purchase. This is different from weather prediction, whatever the weather forecast engine predicts, tomorrow's weather won't be affected. Therefore, when we conduct offline evaluation for recommendation engines, it is possible that the final user behavior is dramatically different from the evaluation result. However, in the evaluation, for simplicity, we have to assume that user behavior is homogenous.</p></li> </ul> <h2 id='evaluation-data-generation' class='header-anchors'>Evaluation Data Generation</h2><h3 id='actual-result' class='header-anchors'>Actual Result</h3><p>In MyRecommendation/src/main/scala/<strong><em>Engine.scala</em></strong>, we define the <code>ActualResult</code> which represents the user rating for validation. It stores the list of ratings in the validation set for a user.</p><div class="highlight scala"><table style="border-spacing: 0"><tbody><tr><td class="gutter gl" style="text-align: right"><pre class="lineno">1 2 3</pre></td><td class="code"><pre><span class="k">case</span> <span class="k">class</span> <span class="nc">ActualResult</span><span class="o">(</span> <span class="n">ratings</span><span class="k">:</span> <span class="kt">Array</span><span class="o">[</span><span class="kt">Rating</span><span class="o">]</span>
