MinHashLSH and BucketedRandomProjectionLSH benchmark #128

MinHashLSH and BucketedRandomProjectionLSH benchmark added.

Future questions:
* Whether we need to improve the way of testing data generation for MinHashLSH ( and add more control param, such as max/min element number in each input set )
* Whether we need to add benchmark for approxNearestNeighbors and approxSimilarityJoin

src/main/scala/com/databricks/spark/sql/perf/mllib/feature/BucketedRandomProjectionLSH.scala

@@ -0,0 +1,35 @@
+package com.databricks.spark.sql.perf.mllib.feature
+
+import org.apache.spark.ml
+import org.apache.spark.ml.PipelineStage
+import org.apache.spark.sql._
+
+import com.databricks.spark.sql.perf.mllib.{BenchmarkAlgorithm, MLBenchContext, TestFromTraining}
+import com.databricks.spark.sql.perf.mllib.OptionImplicits._
+import com.databricks.spark.sql.perf.mllib.data.DataGenerator
+
+/** Object for testing BucketedRandomProjectionLSH performance */
+object BucketedRandomProjectionLSH extends BenchmarkAlgorithm with TestFromTraining {
+
+  override def trainingDataSet(ctx: MLBenchContext): DataFrame = {
+    import ctx.params._
+
+    val df = DataGenerator.generateContinuousFeatures(
+      ctx.sqlContext,
+      numExamples,
+      ctx.seed(),
+      numPartitions,
+      numFeatures
+    )
+    df
+  }
+
+  override def getPipelineStage(ctx: MLBenchContext): PipelineStage = {
+    import ctx.params._
+
+    new ml.feature.BucketedRandomProjectionLSH()
+      .setInputCol("features")
+      .setNumHashTables(numHashTables)
+  }
+
+}

src/main/scala/com/databricks/spark/sql/perf/mllib/feature/MinHashLSH.scala

@@ -0,0 +1,35 @@
+package com.databricks.spark.sql.perf.mllib.feature
+
+import org.apache.spark.ml
+import org.apache.spark.ml.PipelineStage
+import org.apache.spark.sql._
+
+import com.databricks.spark.sql.perf.mllib.OptionImplicits._
+import com.databricks.spark.sql.perf.mllib.data.DataGenerator
+import com.databricks.spark.sql.perf.mllib.{BenchmarkAlgorithm, MLBenchContext, TestFromTraining}
+
+/** Object for testing MinHashLSH performance */
+object MinHashLSH extends BenchmarkAlgorithm with TestFromTraining {
+
+  override def trainingDataSet(ctx: MLBenchContext): DataFrame = {
+    import ctx.params._
+
+    val df = DataGenerator.generateMixedFeatures(
+      ctx.sqlContext,
+      numExamples,
+      ctx.seed(),
+      numPartitions,
+      Array.fill(numFeatures)(2)
+    )
+    df
+  }
+
+  override def getPipelineStage(ctx: MLBenchContext): PipelineStage = {
+    import ctx.params._
+
+    new ml.feature.MinHashLSH()
+      .setInputCol("features")
+      .setNumHashTables(numHashTables)
+  }
+
+}

src/main/scala/com/databricks/spark/sql/perf/results.scala

@@ -130,6 +130,7 @@ class MLParams(
     val maxIter: Option[Int] = None,
     val numClasses: Option[Int] = None,
     val numFeatures: Option[Int] = None,
+    val numHashTables: Option[Int] = Some(1),
     val numInputCols: Option[Int] = None,
     val numItems: Option[Int] = None,
     val numUsers: Option[Int] = None,
@@ -171,6 +172,7 @@ class MLParams(
       maxIter: Option[Int] = maxIter,
       numClasses: Option[Int] = numClasses,
       numFeatures: Option[Int] = numFeatures,
+      numHashTables: Option[Int] = numHashTables,
       numInputCols: Option[Int] = numInputCols,
       numItems: Option[Int] = numItems,
       numUsers: Option[Int] = numUsers,
@@ -185,8 +187,9 @@ class MLParams(
       bucketizerNumBuckets = bucketizerNumBuckets, depth = depth, docLength = docLength,
       elasticNetParam = elasticNetParam, family = family, featureArity = featureArity,
       itemSetSize = itemSetSize, k = k, link = link, maxIter = maxIter,
-      numClasses = numClasses, numFeatures = numFeatures, numInputCols = numInputCols,
-      numItems = numItems, numUsers = numUsers, optimizer = optimizer, regParam = regParam,
+      numClasses = numClasses, numFeatures = numFeatures, numHashTables = numHashTables,
+      numInputCols = numInputCols, numItems = numItems, numUsers = numUsers,
+      optimizer = optimizer, regParam = regParam,
       rank = rank, smoothing = smoothing, tol = tol, vocabSize = vocabSize)
   }
 }

src/main/scala/configs/mllib-small.yaml

@@ -74,6 +74,11 @@ benchmarks:
       optimizer:
         - em
         - online
+  - name: feature.BucketedRandomProjectionLSH
+    params:
+      numExamples: 100
+      numFeatures: 10
+      numHashTables: 1
   - name: feature.Bucketizer
     params:
       numExamples: 100
@@ -83,6 +88,11 @@ benchmarks:
       numExamples: 100
       docLength: 20
       vocabSize: 4
+  - name: feature.MinHashLSH
+    params:
+      numExamples: 100
+      numFeatures: 10
+      numHashTables: 1
   - name: feature.OneHotEncoder
     params:
       numExamples: 100