R Interface for H2O Sparkling Water

An extension package for 'sparklyr' that provides an R interface to H2O Sparkling Water machine learning library (see < https://github.com/h2oai/sparkling-water> for more information).


News

If you are viewing this file on CRAN, please check latest news on GitHub here.

rsparkling 0.2.2 (2017-09-18)

  • Add support for Spark 2.2.* and Sparkling Water 2.2.*
  • Added example of using Sparkling Water w/ YARN here
  • Notes:
    • Added latest rsparkling integration information here
    • Updates to rsparklingREADME

rsparkling 0.2.1 (2017-06-30)

  • Patch fix:
    • Add duke dependency(no.priv.garshol.duke:duke:1.2) to spark_dependency() in package.R, which will help loading the latest H2O (>= 3.10.5.2).
  • Notes:

rsparkling 0.2.0 (2017-03-15)

  • New Features:
    1. options(rsparkling.sparklingwater.location)
      • This feature allows a user to point to a specific Sparkling Water jar before doing a library call to rsparkling. Basic usage is as follows:

        options(rsparkling.sparklingwater.location = "path/to/sparkling_water.jar")
        library(rsparkling) 
        #`spark_connect` will use the local Sparkling Water jar from the above `options()` call. 
        sc <- spark_connect(master = "local", version = "2.0.0") 
    2. h2o_release_table()
      • This new function will display an integration table in your R console as a data.frame. Example output is below:

        > library(rsparkling)
        > h2o_release_table()
           Spark_Version Sparkling_Water_Version H2O_Version H2O_Release_Name
        1            2.1                   2.1.0    3.10.3.2     rel-tverberg
        2            2.0                   2.0.5    3.10.3.2     rel-tverberg
        3            2.0                   2.0.4    3.10.3.2     rel-tverberg
        4            2.0                   2.0.3    3.10.1.2     rel-turnbull
        5            2.0                   2.0.2   3.10.0.10       rel-turing
        6            2.0                   2.0.1   3.10.0.10       rel-turing
        7            2.0                   2.0.0    3.10.0.7       rel-turing
        8            1.6                   1.6.8    3.10.0.7       rel-turing
        9            1.6                   1.6.7    3.10.0.6       rel-turing
        10           1.6                   1.6.6    3.10.0.4       rel-turing
        11           1.6                   1.6.5     3.8.2.6      rel-turchin
        12           1.6                   1.6.4     3.8.2.4      rel-turchin
        13           1.6                   1.6.3     3.8.2.3      rel-turchin
        14           1.6                   1.6.2     3.8.1.3        rel-turan
        15           1.6                   1.6.1     3.8.1.3        rel-turan
           H2O_Release_Patch_Number
        1                         2
        2                         2
        3                         2
        4                         2
        5                        10
        6                        10
        7                         7
        8                         7
        9                         6
        10                        4
        11                        6
        12                        4
        13                        3
        14                        3
        15                        3
  • Notes:
    1. Support for Sparkling Water 2.1.0 has been added to rsparkling.

    2. rsparkling will automatically decipher a Sparkling Water version based on Spark's major version if the user does not provide a specific version of Sparkling Water and a path to a Sparkling Water jar. The logic goes as follows:

      • If Spark 2.1.* is detected, then rsparkling will fetch the latest Sparkling Water for Spark 2.1.*
      • If Spark 2.0.* is detected, then rsparkling will fetch the latest Sparkling Water for Spark 2.0.*
      • If Spark 1.6.* is detected, then rsparkling will fetch the latest Sparkling Water for Spark 1.6.*
        • If the wrong version of H2O is detected for one of the previous versions of Sparkling Water, then the error message will provide a solution to install the correct version.
    3. Install instructions for rsparkling have been refactored and are much more complete at this point (refer to README.md).

    4. Instructions about dealing with firewalls/lack of internet have been added for Spark and Sparkling Water.

    5. A wiki dealing with common problems on Windows OS has been added to the rsparkling repo on Github.

rsparkling 0.1.1 (2017-02-23)

  • Patch fix: Set startH2O to FALSE in h2o.init() command to ensure R only starts H2O if no other H2O instance is detected. This is mainly useful if a user is in a cluster environment(Hadoop,EMR,etc.).

rsparkling 0.1.0 (2017-01-24)

  • Initial release.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("rsparkling")

0.2.12 by Jakub Hava, a month ago


https://github.com/h2oai/sparkling-water/tree/master/r


Report a bug at http://jira.h2o.ai


Browse source code at https://github.com/cran/rsparkling


Authors: Jakub Hava [aut, cre] , Navdeep Gill [aut] , Erin LeDell [aut] , Michal Malohlava [aut] , JJ Allaire [aut] , H2O.ai [cph] , RStudio [cph]


Documentation:   PDF Manual  


Task views: Model Deployment with R


Apache License 2.0 | file LICENSE license


Imports utils, sparklyr, h2o

Suggests dplyr, testthat

System requirements: Java (>6)


See at CRAN