Decision Forest - 'This node requires one or more packages to be downloaded...' - which packages?

Answered

Comments

5 comments

  • Avatar
    Infogix Support

    Hi Drew,

    There's actually a node called: "R Library Package Download" within the Logisitics Library.  

    1. Drag the R Library Package Download onto the canvas
    2. Review the property panel on the right hand side for that node to find the License Agreement
    3. If you agree, change the Agreement property to: I accept the agreement
    4. Run the node

    This node will only need to be run once per installed instance of Dataverse and it will download the required packages for you.

    If you don't have access to that node, the following Help document on Installing CRAN could provide some additional guidance: 

    https://d3sa-preview.infogixsaas.com/docs/dist/help/Default.htm#h-tech-guides/installing-cran.htm%3FTocPath%3DReference%7C_____1

    [Edit: ^ Updated URL]

    0
    Comment actions Permalink
  • Avatar
    Drew Hoffman

    Perfect - working now. Thanks!

    0
    Comment actions Permalink
  • Avatar
    Donald Graham

    1. The link listed above returns a "Page Not Found" 

    2. I receive the following error. How do I proceed or am I not able to use R?

        "Can not execute R Library Package Download due to missing node license feature."

    0
    Comment actions Permalink
  • Avatar
    Adrian Williams

    Hi Donald,

    1. Apologies for the broken link. The information on installing CRAN packages can be found in the product's integrated help in the following topic:

        Reference > Installing CRAN packages 

    It can also be viewed in the online version of the documentation here:

    https://d3sa-preview.infogixsaas.com/docs/dist/help/Default.htm#h-tech-guides/installing-cran.htm%3FTocPath%3DReference%7C_____1

    [Edit: ^ Updated URL]

     

    2. The use of the nodes in the Statistical and Predictive Nodes pack (e.g. the Decision Forest, Linear Regression , Logistic Regression, K-Means Clustering, Time Series Forecast nodes, R Library Package Download node, etc.) requires an additional license. The R Library package Download node is a utility node that automates the download and installation of the prerequisite CRAN packages used by the Statistical and Predictive Nodes. These nodes use the embedded R computation environment rather than leveraging an Open Source R environment.

    You can use Open Source R with Dataverse by downloading and installing the R node, which integrates with an Open Source R environment. No separate license is required to use the R node. You may be interested in reading the following article that describes how you can leverage R to implement Machine Learning:

    https://support.infogix.com/hc/en-us/articles/360018962494-Machine-Learning-with-Data3Sixty-Analyze-and-R

    The article also includes information on how to obtain the R node. The node can also be obtained from the download sections of this forum (login required to access these pages):

    https://support.infogix.com/hc/en-us/sections/360003464673-Downloads

     

    Regards,

    Adrian

    0
    Comment actions Permalink
  • Avatar
    Loretta

    Thanks for your complicated guide. 

     

    (1) Are R node downloading and R installation (to my desktop, like the R software) two different things? The R node in D360 can utilize the packages or the libraries from R software installation? So for running the R node, I have to download R packages to my desktop first, then copy the corresponding downloaded .jar files to Analyze's directory? 

     

    (2) Regarding to your answer: "However, if your script needs to leverage additional functions that are provided by CRAN packages then your R installation will need to be configured with those packages."--------do you mean I can download additional R libraries and save them under D360's directory to leverage these functions as what the statistical and prediction nodes can do?

     

    Sincerely,

    Loretta

     
    0
    Comment actions Permalink

Please sign in to leave a comment.



Powered by Zendesk