Geek Answers Handbook

H2o to R

Now I am studying the h2o package,

  • I installed the h2o package from CRAN and did not execute this code

      ## To import small iris data file from H \: sub: `2` \ O package
 irisPath = system.file ("extdata", "iris.csv", package = "h2o")
 iris.hex = h2o.importFile (localH2O, path = irisPath, key = "iris.hex")

I get the following error:

Error in h2o.importFile (localH2O, path = irisPath, key = "iris.hex"): unused argument (key = "iris.hex")

  1. My second question is: do we have good resources for learning h2o in R, other than this:

http://h2o-release.s3.amazonaws.com/h2o/rel-lambert/5/docs-website/Ruser/rtutorial.html

  1. My third question: I want to know how h2o works in simple words.
+5
source share
4 answers

The reason this code no longer works is because the API changed from H2O 2.0 to H2O 3.0 in 2015. The documents you discovered (probably through a Google search) are from a very old version of H2O 2.0. Updated documents can always be found at http://docs.h2o.ai/h2o/latest-stable/h2o-docs/index.html

+4
source

Answering your error question:

H2O has changed a bit from this documentation. Reading the iris file works as follows:

iris.hex = h2o.importFile(path = irisPath, destination_frame = "iris.hex")

Your second (and third question) is against SO rules. But below you will find a short list of resources:

  • H2O training materials (go to the h2o.ai website) and go to general documentation. You can find all the material presented on the h2o world of 2015. There is also a link to h2o university.
  • Check out your blog. There are several gold nuggets.
  • Read the booklets they have on GBM, GLM, Deep Learning. They contain examples in R and Python.
  • Kaggle. Search scripts / cores for h2o.

Regarding the third question, read their Why H2O Pages.

+3
source

To answer your question about how H2O works, it's hard to put together here. however, in a nutshell, H2O is an open source open source engine available from popular machine learning languages, i.e. R, Python, as well as the Java and Scala programming languages. Enterprise ready means that users can distribute performance across multiple machines depending on an extremely large amount of data. The Java-based kernel has a built-in implementation of several algorithms, and any language interface passes through the interpreter with the H2O core, which can be a distributed cluster to create models and evaluate results. There is a lot between them, so I would suggest a link to the link below to learn more about H2O architecture and execution using various supported languages:

http://docs.h2o.ai/h2o/latest-stable/h2o-docs/architecture.html

+1
source

You can more compute the implementation of H2O in R, starting from the installation and ending with the implementation of the h2o computer learning library in R. Follow this link . It will also help you implement h2o machine learning on top of the SparkR infrastructure .

If you want to get an idea of ​​the h2o working prototype from a very simple one , then this link . It provides the basic taste of a working prototype using code (a quick learning tutorial).

In addition to the points above, it also covers the following key points:

  • How to convert an H2O data frame to an R and Spark data frame and vice versa
  • What are the pros and cons between the SparkMLlib machine library and H2O.
  • What are the strengths of h2o compared to other ML libraries.
  • How to apply ML algorithm to R and Spark data frames, etc.
0
source

More articles:


All Articles