Executing an R Script

Question

Executing an R Script

Steve Jones - SSC Editor

SSC Guru

Points: 734418
More actions
October 13, 2016 at 11:55 pm

#72654

Comments posted to this topic are about the item Executing an R Script

Viewing 12 posts - 1 through 11 (of 11 total)

You must be logged in to reply to this topic. Login to reply

Iulian -207023 SSCertifiable Points: 7639 More actions · Answer 1

Nice one, but I thought there is a way to pass just the script's name.

The same way there is catalog for SSIS and a reporting DB for SSRS, it would be nice to be able to create an R library on SQL Server.

Ta

Iulian

paul s-306273 SSChampion Points: 10727 More actions · Answer 2

If you got QOTD right for Oct 7th, you probably got this right too!

George Vobr SSCrazy Eights Points: 9914 More actions · Answer 3

paul s-306273 (10/14/2016)
If you got QOTD right for Oct 7th, you probably got this right too!

+1 😉

Yes, easy one. Thanks Steve for this question.

Ed Wagner SSC Guru Points: 287024 More actions · Answer 4

George Vobr (10/14/2016)
paul s-306273 (10/14/2016)
If you got QOTD right for Oct 7th, you probably got this right too!
+1 😉
Yes, easy one. Thanks Steve for this question.

I didn't know the date, but remembered the question. Thanks, Steve.

Tally Tables - Performance Personified
String Splitting with True Performance
Best practices on how to ask questions

sjimmo SSChampion Points: 11139 More actions · Answer 5

I must have been sleeping on Oct 7, and accidently got it right. But I too thought you could pass a script name to R

Steve Jimmo
Sr DBA
“If we ever forget that we are One Nation Under God, then we will be a Nation gone under." - Ronald Reagan

sknox SSChampion Points: 12388 More actions · Answer 6

You can use source in the text of your R script to bring in a file. Cf http://stackoverflow.com/questions/3502925/include-files-r

Steve Jones - SSC Editor SSC Guru Points: 734418 More actions · Answer 7

Iulian -207023 (10/13/2016)
Nice one, but I thought there is a way to pass just the script's name.
The same way there is catalog for SSIS and a reporting DB for SSRS, it would be nice to be able to create an R library on SQL Server.
Ta
Iulian

The script can be a variable. Putting this separate makes it easier to read.

Revenant SSC-Forever Points: 42467 More actions · Answer 8

Steve Jones - SSC Editor (10/14/2016)
Iulian -207023 (10/13/2016)
Nice one, but I thought there is a way to pass just the script's name.
The same way there is catalog for SSIS and a reporting DB for SSRS, it would be nice to be able to create an R library on SQL Server.
Ta
Iulian
The script can be a variable. Putting this separate makes it easier to read.

This is more flexible than it seems. You can put R script into SQL code and in that R you can call SQL that calls another R script, there is no stated limit how deep you can go.

You can keep your R scripts in a simple table, and more importantly, you can store your trained models in a table and pass them to R just for evaluation, which is devilishly fast. Nice symbiosis.

(Yes, this hits my hot button. 🙂 )

Iulian -207023 SSCertifiable Points: 7639 More actions · Answer 9

Revenant (10/14/2016)
Steve Jones - SSC Editor (10/14/2016)
Iulian -207023 (10/13/2016)
Nice one, but I thought there is a way to pass just the script's name.
The same way there is catalog for SSIS and a reporting DB for SSRS, it would be nice to be able to create an R library on SQL Server.
Ta
Iulian
The script can be a variable. Putting this separate makes it easier to read.
This is more flexible than it seems. You can put R script into SQL code and in that R you can call SQL that calls another R script, there is no stated limit how deep you can go.
You can keep your R scripts in a simple table, and more importantly, you can store your trained models in a table and pass them to R just for evaluation, which is devilishly fast. Nice symbiosis.
(Yes, this hits my hot button. 🙂 )

Yes, it make sense to keep the R script on a variable or to load it from a table.

T.a.

Iulian

Iulian -207023 SSCertifiable Points: 7639 More actions · Answer 10

Iulian -207023 (10/17/2016)
Revenant (10/14/2016)
Steve Jones - SSC Editor (10/14/2016)
Iulian -207023 (10/13/2016)
Nice one, but I thought there is a way to pass just the script's name.
The same way there is catalog for SSIS and a reporting DB for SSRS, it would be nice to be able to create an R library on SQL Server.
Ta
Iulian
The script can be a variable. Putting this separate makes it easier to read.
This is more flexible than it seems. You can put R script into SQL code and in that R you can call SQL that calls another R script, there is no stated limit how deep you can go.
You can keep your R scripts in a simple table, and more importantly, you can store your trained models in a table and pass them to R just for evaluation, which is devilishly fast. Nice symbiosis.
(Yes, this hits my hot button. 🙂 )
Yes, it make sense to keep the R script on a variable or to load it from a table.
T.a.
Iulian

Very powerful because it sounds like a delegate, treating a function as a variable, sending functions as parameters.

There is support for this sort of things in T-SQL using EXECUTE, but this R seems to add more functionality through the packages it comes with.

By the way, what is the execution context for R? Is there a console or something to see what packages are installed or how can I install more packages ?

T.a.

Iulian

Koen Verbeeck SSC Guru Points: 259085 More actions · Answer 11

Easy one, thanks.

Need an answer? No, you need a question
My blog at https://sqlkover.com.
MCSE Business Intelligence - Microsoft Data Platform MVP