diff options
Diffstat (limited to 'core/src/benchmark/scala/com/rockymadden/stringmetric/similarity')
10 files changed, 483 insertions, 0 deletions
diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala new file mode 100755 index 0000000..5df27fb --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class DiceSorensenMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala new file mode 100755 index 0000000..3d04074 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class HammingMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + HammingMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + HammingMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + HammingMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + HammingMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala new file mode 100755 index 0000000..ddeef7a --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaccardMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaccardMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaccardMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaccardMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaccardMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala new file mode 100755 index 0000000..47baa6d --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaroMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaroMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaroMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaroMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaroMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala new file mode 100755 index 0000000..c11a01c --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaroWinklerMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaroWinklerMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaroWinklerMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaroWinklerMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaroWinklerMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala new file mode 100755 index 0000000..a8460e2 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class LevenshteinMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + LevenshteinMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + LevenshteinMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + LevenshteinMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + LevenshteinMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala new file mode 100755 index 0000000..8ba1bb0 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala @@ -0,0 +1,51 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class NGramMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + @Param(Array("2", "3")) + var n: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + NGramMetric(n).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + NGramMetric(n).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + NGramMetric(n).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + NGramMetric(n).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala new file mode 100755 index 0000000..89207f2 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class OverlapMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + OverlapMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + OverlapMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + OverlapMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + OverlapMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala new file mode 100755 index 0000000..86196ab --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class RatcliffObershelpMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala new file mode 100755 index 0000000..837ce01 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class WeightedLevenshteinMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(string1, string1) + } +} |