Skip to main content
Announcements
Introducing Qlik Answers: A plug-and-play, Generative AI powered RAG solution. READ ALL ABOUT IT!
cancel
Showing results for 
Search instead for 
Did you mean: 
simonegabrielli
Contributor III
Contributor III

Qlik Sense and R integration: R library

Hi everyone, in the installation PDF guide are indicated the following  R library:

 

install.packages("jsonlite", lib = "C:\\R\\R-3.6.1\\library")

install.packages("ChainLadder", lib = "C:\\R\\R-3.6.1\\library")

install.packages("forecast", lib = "C:\\R\\R-3.6.1\\library")

install.packages("rpart", lib = "C:\\R\\R-3.6.1\\library")

install.packages("d3r", lib = "C:\\R\\R-3.6.1\\library")

install.packages("partykit", lib = "C:\\R\\R-3.6.1\\library")

install.packages("tseries", lib = "C:\\R\\R-3.6.1\\library")

install.packages("colorspace", lib = "C:\\R\\R-3.6.1\\library")

 

Can I use other R libraries in Qlik Sense Desktop, and call them in the Load Script Editor ?

I tried to use the library (Benchmarking) but does not work. In SSEtoRserve the error is: Cannot decode an Sexp because the type is not recognized: 48

the script is the following: 

SSELoad:
Load
INPUT1,
INPUT2,
OUTPUT1,
OUTPUT2,
OUTPUT3,
OUTPUT4,

R.ScriptEval('cbind(q$INPUT1, q$INPUT2);', INPUT1, INPUT2) as X,

R.ScriptEval('cbind(q$OUTPUT1, q$OUTPUT2, q$OUTPUT3, q$OUTPUT4);', OUTPUT1, OUTPUT2, OUTPUT3, OUTPUT4) as Y

Resident DEALOAD;

SSEload2:

LOAD
X,
Y,

R.ScriptEval('library(Benchmarking); dea(q$X, q$Y);', X, Y) AS DEA

Resident SSELoad;

 

Qlik can not work with this R library, or there is an error in the script ?

1 Reply
maitegarcia
Contributor
Contributor

Hello,

There're no possible answers on this topic ?

I encounter the same type of error when I call a glm model created whith H2o library :

R.ScriptEval('
library(h2o);
h2o.init();
q <- as.h2o(q);
gbm_path = load("D:/Users/maite.garcia/Desktop/path.RData");
gbm = h2o.loadModel(path);
res <- h2o.predict(object = gbm, newdata = q);
res$predict;
',
Avg(varX_1) as varX_1,
Avg(varX_2) as varX_2,
Avg(varX_3) as varX_3)

Thanks for all !