Connecting Knime to R

Hi all,

I have downloaded the R to the KNIME plug-in. I have a script in R that I would like to excute automatically between KNIME nodes, for example. What would be the suitable node for that?

 

1- READ CSV NODE --> DECESION TREE LEARNER NODE --> DECESION TREE PREDICTOR NODE

2- Run my R script

3- Continue running the remaining nodes

 

Note: the "script in R" doesn't have direct input or output from the previous decision tree predictions, but I would like to run them in order.

Many thanks,

Error404

You can always put a node in-between two others, either using data connections or the variables ports. Those can be used to define an execution order on nodes without explicit variable ports. You can enable those special ports in the node contect menu.

Hi Gabriel,

Let's assume a simple case here to avoid confusion, can we try a little example please?

I would like to perform the following R command without having to go to the R program and excute it.

 

file1 <- read.csv("\\\\myfile1.csv",sep=",")

file2 <- read.csv("\\\\myfile2.csv",sep=",")

file3 <- match(file1,file2)

write.csv(file3, "my_location\\ file3.csv")

 

Would that be possible in Knime?

All the R nodes asks for "input" and has a line at the beggining and show this line

R<-R

Thanks :)

 

 

 


 

Hello, 

Have you looked at the Public Examples server?  There are a couple of workflows in there that demonstrate the usage of the R-nodes. 

Regards,

Aaron

Hello Aaron,

 

Thanks for the hint, I have downloaded the examples, but even these examples give me the error "read.csv not found" when I try to excute them. I am still doing things in R manually, although they are part of my workflow :(