spark-knn-graphs (homepage)

Spark algorithms for building and processing k-nn graphs

@tdebatty / (1)

Spark algorithms for building k nearest neighbors (k-nn) graphs. The package includes implementations of
 - Brute force graph building;
 - NN-Descent graph building;
 - Fast online graph building;
 - LSH SuperBit graph building;
 - Online graph building.


Tags

  • 1|machine learning
  • 1|graph

How to

Include this package in your Spark Applications using:

spark-shell, pyspark, or spark-submit

> $SPARK_HOME/bin/spark-shell --packages tdebatty:spark-knn-graphs:0.13

sbt

If you use the sbt-spark-package plugin, in your sbt build file, add:

spDependencies += "tdebatty/spark-knn-graphs:0.13"

Otherwise,

resolvers += "Spark Packages Repo" at "https://repos.spark-packages.org/"

libraryDependencies += "tdebatty" % "spark-knn-graphs" % "0.13"

Maven

In your pom.xml, add:
<dependencies>
  <!-- list of dependencies -->
  <dependency>
    <groupId>tdebatty</groupId>
    <artifactId>spark-knn-graphs</artifactId>
    <version>0.13</version>
  </dependency>
</dependencies>
<repositories>
  <!-- list of other repositories -->
  <repository>
    <id>SparkPackagesRepo</id>
    <url>https://repos.spark-packages.org/</url>
  </repository>
</repositories>

Releases

Version: 0.13 ( a4bfc5 | zip | jar ) / Date: 2016-02-17 / License: MIT

Version: 0.12 ( 2e639c | zip | jar ) / Date: 2016-02-08 / License: MIT

Spark Scala/Java API compatibility: - 100% , - 83% , - 100% , - 87%

Version: 0.11 ( 734ee0 | zip | jar ) / Date: 2016-01-26 / License: MIT

Spark Scala/Java API compatibility: - 83% , - 100% , - 87% , - 100%

Version: 0.9 ( 0aa35f | zip | jar ) / Date: 2016-01-12 / License: MIT

Spark Scala/Java API compatibility: - 100% , - 83% , - 100% , - 83%

Version: 0.7 ( 4bbe1d | zip | jar ) / Date: 2015-12-13 / License: LGPL-3.0

Version: 0.6 ( c3a0ab | zip | jar ) / Date: 2015-12-01 / License: LGPL-3.0

Version: 0.5 ( 064cf2 | zip | jar ) / Date: 2015-11-27 / License: MIT

Version: 0.4 ( f8b99a | zip | jar ) / Date: 2015-07-15 / License: MIT

Spark Scala/Java API compatibility: - 80% , - 100% , - 100%