diff options
| author | Matei Alexandru Zaharia <matei@apache.org> | 2014-05-03 21:55:05 +0000 |
|---|---|---|
| committer | Matei Alexandru Zaharia <matei@apache.org> | 2014-05-03 21:55:05 +0000 |
| commit | 60c97ec11599b2339cea8bd0e005d29bfe8f5d6c (patch) | |
| tree | 8249ea7b1cb32f881b520f3dae33e28872975b40 /examples.md | |
| parent | 1d51eb75ca54a02b03f61d75ac0583221b0e3869 (diff) | |
| download | spark-website-60c97ec11599b2339cea8bd0e005d29bfe8f5d6c.tar.gz spark-website-60c97ec11599b2339cea8bd0e005d29bfe8f5d6c.tar.bz2 spark-website-60c97ec11599b2339cea8bd0e005d29bfe8f5d6c.zip | |
show Python examples first
Diffstat (limited to 'examples.md')
| -rw-r--r-- | examples.md | 124 |
1 files changed, 62 insertions, 62 deletions
diff --git a/examples.md b/examples.md index 71156ef80..ec381c124 100644 --- a/examples.md +++ b/examples.md @@ -19,12 +19,24 @@ previous ones, and <em>actions</em>, which kick off a job to execute on a cluste In this example, we search through the error messages in a log file: <ul class="nav nav-tabs"> - <li class="lang-tab lang-tab-scala active"><a href="#">Scala</a></li> + <li class="lang-tab lang-tab-python active"><a href="#">Python</a></li> + <li class="lang-tab lang-tab-scala"><a href="#">Scala</a></li> <li class="lang-tab lang-tab-java"><a href="#">Java</a></li> - <li class="lang-tab lang-tab-python"><a href="#">Python</a></li> </ul> <div class="tab-content"> - <div class="tab-pane tab-pane-scala active"> + <div class="tab-pane tab-pane-python active"> + <div class="code code-tab"> + file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> + errors = file.<span class="sparkop">filter</span>(<span class="closure">lambda line: "ERROR" in line</span>)<br> + <span class="comment"># Count all the errors</span><br> + errors.<span class="sparkop">count</span>()<br> + <span class="comment"># Count errors mentioning MySQL</span><br> + errors.<span class="sparkop">filter</span>(<span class="closure">lambda line: "MySQL" in line</span>).<span class="sparkop">count</span>()<br> + <span class="comment"># Fetch the MySQL errors as an array of strings</span><br> + errors.<span class="sparkop">filter</span>(<span class="closure">lambda line: "MySQL" in line</span>).<span class="sparkop">collect</span>()<br> + </div> + </div> + <div class="tab-pane tab-pane-scala"> <div class="code code-tab"> <span class="keyword">val</span> file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> <span class="keyword">val</span> errors = file.<span class="sparkop">filter</span>(<span class="closure">line => line.contains("ERROR")</span>)<br> @@ -54,18 +66,6 @@ In this example, we search through the error messages in a log file: }</span>).<span class="sparkop">collect</span>();<br> </div> </div> - <div class="tab-pane tab-pane-python"> - <div class="code code-tab"> - file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> - errors = file.<span class="sparkop">filter</span>(<span class="closure">lambda line: "ERROR" in line</span>)<br> - <span class="comment"># Count all the errors</span><br> - errors.<span class="sparkop">count</span>()<br> - <span class="comment"># Count errors mentioning MySQL</span><br> - errors.<span class="sparkop">filter</span>(<span class="closure">lambda line: "MySQL" in line</span>).<span class="sparkop">count</span>()<br> - <span class="comment"># Fetch the MySQL errors as an array of strings</span><br> - errors.<span class="sparkop">filter</span>(<span class="closure">lambda line: "MySQL" in line</span>).<span class="sparkop">collect</span>()<br> - </div> - </div> </div> <p>The red code fragments are function literals (closures) that get passed automatically to the cluster. The blue ones are Spark operations.</p> @@ -75,24 +75,24 @@ In this example, we search through the error messages in a log file: <p>Spark can <em>cache</em> datasets in memory to speed up reuse. In the example above, we can load just the error messages in RAM using:</p> <ul class="nav nav-tabs"> - <li class="lang-tab lang-tab-scala active"><a href="#">Scala</a></li> + <li class="lang-tab lang-tab-python active"><a href="#">Python</a></li> + <li class="lang-tab lang-tab-scala"><a href="#">Scala</a></li> <li class="lang-tab lang-tab-java"><a href="#">Java</a></li> - <li class="lang-tab lang-tab-python"><a href="#">Python</a></li> </ul> <div class="tab-content"> - <div class="tab-pane tab-pane-scala active"> + <div class="tab-pane tab-pane-python active"> <div class="code code-tab"> errors.<span class="sparkop">cache</span>() </div> </div> - <div class="tab-pane tab-pane-java"> + <div class="tab-pane tab-pane-scala"> <div class="code code-tab"> - errors.<span class="sparkop">cache</span>(); + errors.<span class="sparkop">cache</span>() </div> </div> - <div class="tab-pane tab-pane-python"> + <div class="tab-pane tab-pane-java"> <div class="code code-tab"> - errors.<span class="sparkop">cache</span>() + errors.<span class="sparkop">cache</span>(); </div> </div> </div> @@ -105,12 +105,21 @@ In this example, we search through the error messages in a log file: <p>In this example, we use a few more transformations to build a dataset of (String, Int) pairs called <code>counts</code> and then save it to a file.</p> <ul class="nav nav-tabs"> - <li class="lang-tab lang-tab-scala active"><a href="#">Scala</a></li> + <li class="lang-tab lang-tab-python active"><a href="#">Python</a></li> + <li class="lang-tab lang-tab-scala"><a href="#">Scala</a></li> <li class="lang-tab lang-tab-java"><a href="#">Java</a></li> - <li class="lang-tab lang-tab-python"><a href="#">Python</a></li> </ul> <div class="tab-content"> - <div class="tab-pane tab-pane-scala active"> + <div class="tab-pane tab-pane-python active"> + <div class="code code-tab"> + file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> + counts = file.<span class="sparkop">flatMap</span>(<span class="closure">lambda line: line.split(" ")</span>) \<br> + .<span class="sparkop">map</span>(<span class="closure">lambda word: (word, 1)</span>) \<br> + .<span class="sparkop">reduceByKey</span>(<span class="closure">lambda a, b: a + b</span>)<br> + counts.<span class="sparkop">saveAsTextFile</span>(<span class="string">"hdfs://..."</span>) + </div> + </div> + <div class="tab-pane tab-pane-scala"> <div class="code code-tab"> <span class="keyword">val</span> file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> <span class="keyword">val</span> counts = file.<span class="sparkop">flatMap</span>(<span class="closure">line => line.split(" ")</span>)<br> @@ -134,15 +143,6 @@ In this example, we search through the error messages in a log file: counts.<span class="sparkop">saveAsTextFile</span>(<span class="string">"hdfs://..."</span>); </div> </div> - <div class="tab-pane tab-pane-python"> - <div class="code code-tab"> - file = spark.textFile(<span class="string">"hdfs://..."</span>)<br> - counts = file.<span class="sparkop">flatMap</span>(<span class="closure">lambda line: line.split(" ")</span>) \<br> - .<span class="sparkop">map</span>(<span class="closure">lambda word: (word, 1)</span>) \<br> - .<span class="sparkop">reduceByKey</span>(<span class="closure">lambda a, b: a + b</span>)<br> - counts.<span class="sparkop">saveAsTextFile</span>(<span class="string">"hdfs://..."</span>) - </div> - </div> </div> <h3>Estimating Pi</h3> @@ -150,12 +150,22 @@ In this example, we search through the error messages in a log file: <p>Spark can also be used for compute-intensive tasks. This code estimates <span style="font-family: serif; font-size: 120%;">π</span> by "throwing darts" at a circle. We pick random points in the unit square ((0, 0) to (1,1)) and see how many fall in the unit circle. The fraction should be <span style="font-family: serif; font-size: 120%;">π / 4</span>, so we use this to get our estimate.</p> <ul class="nav nav-tabs"> - <li class="lang-tab lang-tab-scala active"><a href="#">Scala</a></li> + <li class="lang-tab lang-tab-python active"><a href="#">Python</a></li> + <li class="lang-tab lang-tab-scala"><a href="#">Scala</a></li> <li class="lang-tab lang-tab-java"><a href="#">Java</a></li> - <li class="lang-tab lang-tab-python"><a href="#">Python</a></li> </ul> <div class="tab-content"> - <div class="tab-pane tab-pane-scala active"> + <div class="tab-pane tab-pane-python active"> + <div class="code code-tab"> + <span class="keyword">def</span> sample(p):<br> + x, y = random(), random()<br> + <span class="keyword">return</span> 1 <span class="keyword">if</span> x*x + y*y < 1 <span class="keyword">else</span> 0<br><br> + count = spark.parallelize(xrange(0, NUM_SAMPLES)).<span class="sparkop">map</span>(<span class="closure">sample</span>) \<br> + .<span class="sparkop">reduce</span>(<span class="closure">lambda a, b: a + b</span>)<br> + print <span class="string">"Pi is roughly %f"</span> % (4.0 * count / NUM_SAMPLES)<br> + </div> + </div> + <div class="tab-pane tab-pane-scala"> <div class="code code-tab"> <span class="keyword">val</span> count = spark.parallelize(1 to NUM_SAMPLES).<span class="sparkop">map</span>(<span class="closure">i =><br> val x = Math.random()<br> @@ -177,16 +187,6 @@ In this example, we search through the error messages in a log file: System.out.println(<span class="string">"Pi is roughly "</span> + 4 * count / NUM_SAMPLES);<br> </div> </div> - <div class="tab-pane tab-pane-python"> - <div class="code code-tab"> - <span class="keyword">def</span> sample(p):<br> - x, y = random(), random()<br> - <span class="keyword">return</span> 1 <span class="keyword">if</span> x*x + y*y < 1 <span class="keyword">else</span> 0<br><br> - count = spark.parallelize(xrange(0, NUM_SAMPLES)).<span class="sparkop">map</span>(<span class="closure">sample</span>) \<br> - .<span class="sparkop">reduce</span>(<span class="closure">lambda a, b: a + b</span>)<br> - print <span class="string">"Pi is roughly %f"</span> % (4.0 * count / NUM_SAMPLES)<br> - </div> - </div> </div> <h3>Logistic Regression</h3> @@ -194,12 +194,24 @@ In this example, we search through the error messages in a log file: <p>This is an iterative machine learning algorithm that seeks to find the best hyperplane that separates two sets of points in a multi-dimensional feature space. It can be used to classify messages into spam vs non-spam, for example. Because the algorithm applies the same MapReduce operation repeatedly to the same dataset, it benefits greatly from caching the input in RAM across iterations.</p> <ul class="nav nav-tabs"> - <li class="lang-tab lang-tab-scala active"><a href="#">Scala</a></li> + <li class="lang-tab lang-tab-python active"><a href="#">Python</a></li> + <li class="lang-tab lang-tab-scala"><a href="#">Scala</a></li> <li class="lang-tab lang-tab-java"><a href="#">Java</a></li> - <li class="lang-tab lang-tab-python"><a href="#">Python</a></li> </ul> <div class="tab-content"> - <div class="tab-pane tab-pane-scala active"> + <div class="tab-pane tab-pane-python active"> + <div class="code code-tab"> + points = spark.textFile(...).<span class="sparkop">map</span>(parsePoint).<span class="sparkop">cache</span>()<br> + w = numpy.random.ranf(size = D) <span class="comment"># current separating plane</span><br> + <span class="keyword">for</span> i <span class="keyword">in</span> range(ITERATIONS):<br> + gradient = points.<span class="sparkop">map</span>(<span class="closure"><br> + lambda p: (1 / (1 + exp(-p.y*(w.dot(p.x)))) - 1) * p.y * p.x<br> + </span>).<span class="sparkop">reduce</span>(<span class="closure">lambda a, b: a + b</span>)<br> + w -= gradient<br> + print <span class="string">"Final separating plane: %s"</span> % w<br> + </div> + </div> + <div class="tab-pane tab-pane-scala"> <div class="code code-tab"> <span class="keyword">val</span> points = spark.textFile(...).<span class="sparkop">map</span>(parsePoint).<span class="sparkop">cache</span>()<br> <span class="keyword">var</span> w = Vector.random(D) <span class="comment">// current separating plane</span><br> @@ -231,18 +243,6 @@ In this example, we search through the error messages in a log file: System.out.println(<span class="string">"Final separating plane: "</span> + w);<br> </div> </div> - <div class="tab-pane tab-pane-python"> - <div class="code code-tab"> - points = spark.textFile(...).<span class="sparkop">map</span>(parsePoint).<span class="sparkop">cache</span>()<br> - w = numpy.random.ranf(size = D) <span class="comment"># current separating plane</span><br> - <span class="keyword">for</span> i <span class="keyword">in</span> range(ITERATIONS):<br> - gradient = points.<span class="sparkop">map</span>(<span class="closure"><br> - lambda p: (1 / (1 + exp(-p.y*(w.dot(p.x)))) - 1) * p.y * p.x<br> - </span>).<span class="sparkop">reduce</span>(<span class="closure">lambda a, b: a + b</span>)<br> - w -= gradient<br> - print <span class="string">"Final separating plane: %s"</span> % w<br> - </div> - </div> </div> <p>Note that the current separating plane, <code>w</code>, gets shipped automatically to the cluster with every <code>map</code> call.</p> |