How to install a particular R package on SQL Server 2016.Package like 'ggplot2'.There is the command in R is 'Install.Packages("ggplot2")'.But how we can install this package on SQL Server 2016.
Have a look at the MSDN documentation.
You can also look at a blog post I wrote about how to install Facebook Prophet on SQL Server R Services.
Hope this helps!
Niels
I kept getting errors regarding package management for SQL Server 2017 "Error: The package management feature is not enabled for the current user or not supported on SQL Server version 14".
What finally worked for me was to go to the R version installed with SQL, then install the R packages as I normally would through R, but for the System Library instead of the User Library:
install.packages("tidyverse", lib="C:\\Program Files\\Microsoft SQL Server\\MSSQL14.MSSQLSERVER\\R_SERVICES\\library")
You need to run R as admin to execute the code above, otherwise you will get an error that the folder is not writeable.
If you run in R the command:
.libPaths()
it will show you the user path and the system path. The packages you install normally forces them to go to the user path. I suspect SQL Server can't find it if it's in the user path.
Related
According to this link, SQL Server machine learning service's highest R version is 3.3. Does anyone know when one can upgrade to version 3.5 or later?
Or is my only option to switch to sql server 2019?
Thanks.
Well cs0815, let's take a look.
According to your post, you tagged this with sql-server-2017, so I'll be basing this on that version.
After reading through your link and this one, I was thinking there might be two options:
Hack a new version of R into the R-Services module of SQL server.
Setup an R-Client with version 3.5
However, it looks like R-Client is for a different use case than you might need so we'll focus on an attempt to replace R in R-Services with a higher version directly.
After installing a new instance of 2017 with the following features:
R Services natively installs to:
"C:\Program Files\Microsoft SQL Server\MSSQL14.MSSQLSERVER\R_SERVICES"
Using WinDiff, I pulled the following comparisons vs CRAN R 3.5.3 and Microsoft Open R 3.5.3 and not surprisingly the Microsoft Open R is much closer.
Full diffs:
R_Services vs CRAN R
R_Services vs MS OPEN R
Personally, I'm not at the level to understand all the differences in R Core / runtime between R 3.3 (R_Services) and R 3.5.3 so of main interest to me is the library diff.
What exists in R_Services which does not in Open R is what I'm assuming to be the core of what makes R_Services work with SQL Server so that will be where we need to fill in those gaps with our "replacement" R_Services but with v3.5.3.
Libraries which are completely missing - not just added docs or some small difference:
compatibilityapi
dorsr
microsoftml
mrsdeploy
olapr
revopemar
revoscaler
revotreeview
sqlrutils
And that information leads to us to the Microsoft MLS Docs "R Packages" resource:
https://learn.microsoft.com/en-us/machine-learning-server/r-reference/introducing-r-server-r-package-reference
Which, no, it looks like those packages are not publicly available through MRAN, CRAN, or even directly via devtools from github.
Still, searching brought me onto this possibility:
https://learn.microsoft.com/en-us/sql/advanced-analytics/install/upgrade-r-and-python?view=sql-server-2017#offline-binding-no-internet-access
Download after signing up with Visual Studio Dev Services:
https://learn.microsoft.com/en-us/machine-learning-server/install/machine-learning-server-windows-install#download-machine-learning-server-installer
So - binding this version of MLS to our existing instance:
After that installation is fully complete, let's check and see what's available on our server!
So first things first, this being a new instance of SQL Server - Set to enable external scripts:
sp_configure 'external scripts enabled', 1;
RECONFIGURE WITH OVERRIDE;
And then to bring this soup to nuts:
EXECUTE sp_execute_external_script #language = N'R'
, #script = N'
print(R.Version())
';
output below:
$language
[1] "R"
$version.string
[1] "R version 3.5.2 (2018-12-20)"
$nickname
[1] "Eggshell Igloo"
TLDR: Yes - download the MLS 9.4.7 packages from Visual Studio Dev Services and install on server host.
I am working on a machine without admin rights. I use sql developer to connect to an internal database. I would like to connect via R also.
Is there any way I can do this, without admin rights? Some solutions require me to set up a systemDNS - which I can not do.
Other requires me to install jvm.dll
My environment: Windows7, sqldeveloper, connection method is via TNS file.
Connecting to SQL Developer via R is far more difficult than other databases I've encountered. It's important that you have jdbc6.jar installed on your machine, and that you know the file path to where it was installed. Installing the jar file does not require admin rights. You can install the jar file from Oracle's website.
I use the RJDBC package to connect like so:
library(RJDBC)
jdbcDriver <- JDBC("oracle.jdbc.OracleDriver", classPath = "file path to where ojdbc6.jar is installed on your computer")
jdbcConnection <- dbConnect(jdbcDriver, "jdbc:oracle:thin:#YOUR_SERVER","YOUR_USERNAME","YOUR_PASSWORD")
You can then test the connection with a number of commands; I typically use:
dbListTables(jdbcConnection)
Another favorite of mine is to use dbplyr for dplyr-like functions when working with databases:
library(dbplyr)
tbl(jdbcConnection, "SAMPLE_TABLE_NAME")
The resulting output will be the data from the queried table in tibble form.
You can set the environment variables in your R session.
Sys.setenv(OCI_LIB64="/Path/to/instantclient",OCI_INC="/Path/to/instantclient/sdk/include")
You can put this in the file .Rprofile in your home directory, and RStudio will run it each time you begin a new session. Once you have this in .Rprofile you should be able to install ROracle.
Hi I'm new in infrastructure and still learning. I would like to ask a question:
If there are 2 servers (both using Red Hat), 1 for RStudio and 1 for MiniCran. MiniCran server has full access to internet, but not for RStudio. RStudio Server can ping to MiniCran Server. But, when I tried to install packages in RStudio Server from MiniCran Server (All packages are already in MiniCran Server.). The error is Unable to to access index repository http://xxx.xx.xx/minicran/src/contrib.
But if I tried to install in MiniCran server using local file, it can be installed.
Is that related firewall or anything else?
Thanks so much in advance for anyone who wants to help give a suggest or answer.
RAndFriends, which includes all the items needed to run RExcel, includes just R 2.15.2.
I am currently using the latest version of R but rcom 2.3.1 and rscproxy 2.0.5, which I am currently using, do not allow me to start an R server within Excel.
Setting foreground R server whithin Excel returns a fatal error and R rejects any connection with Excel via rcom.
I get two error messages:
R Server not available
There seems to be no R process connected to Excel
The main difference with a running RExcel session is that in the latter you may see rscproxy and rcom to be loaded when you start a RExcel session.
Is there anyone who is currently using RExcel with R 3.0.1 who can explain me how he succeeded in running it (step by step)?
Found it on statconn's Wiki section.
Assuming you have a suitable version of R installed, the following steps are necessary to install RExcel and the infrastructure. You need to be logged into Windows with administrator privileges to do this!
You also need to follow these instructions if you upgrade R, i.e. you install a new release of R after you have installed RExcel.
Download the statconn DCOM server and execute the program you downloaded.
Start R as administrator (on Windows 7 you need to right-click the R icon and click the corresponding item).
In R, run the following commands (you must start R as administrator to do this).
Commands:
install.packages(c("rscproxy", "rcom"), repos = "http://rcom.univie.ac.at/download", lib = .Library)
library(rcom)
comRegisterRegistry()
Now you have rcom installed, but RExcel is not installed yet.
To install RExcel: download the RExcel installer and run this installation program. Installing RExcel this way will set the background server of R as the default R server for RExcel. You can change this in the configuration settings in R. If you want to set the foreground server as the default site wide server, there is an appropriate option in one of the dialogs of the installation.
The RExcel installer modifies one of the configuration files of R, the file Rprofile.site, usually found at a location like C:\Program Files\R\R-2.13.1\etc\Rprofile.site.
If you do not install RExcel and want the package rcom to be loaded into R each time you start it, you have to add the line
library(rcom)
to Rprofile.site. You have to start your editor as administrator to be able to modify this file.
I thint that focus of those instuctions are on "Assuming you have a suitable version of R installed". R 3.0.1 do not work with RExcel.
I have noticed all development of Rmetrics products have ceased development since Diethelm Wuertz untimely passing, some of his associates on the team maintain it, but further development stopped abruptly after Diethelm Wuertz the project leader and main inspiration behind the team, died in a car accident in 2015, that is how long it has been since any serious development has occurred on Rexcel and all the range of Rmetric products, it is a real tragedy they are still cutting edge 6 years after Diethelm's passing, he would be sad to see his legacy slowly die. I am looking at xlwings and converting not too difficult code to python for speed and power, as far as many great R libraries go, I have not the time to reinvent the wheel within python, it is not a labour of love for me.
Installing the RODBC package on Ubuntu is a bit of a kludge. First I learned to install the following:
$ sudo apt-get install r-cran-rodbc
That wasn't good enough as the package was still looking for header files. I solved this issue by:
$ sudo apt-get install unixodbc-dev
Good, RODBC installed properly on the Ubuntu machine. But when I try to run the following script:
## import excel file from Dropbox
require("RODBC")
channel <- odbcConnectExcel("~/Dropbox/DATA/SAMPLE/petro.xls")
petro <- sqlFetch (channel, "weekly")
odbcClose(channel)
str(petro)
head(petro)
I get an error thrown that function odbcConnectExcel not found. I checked the case of each letter, making sure it was not a simple typo. Nope. Then I ran this same script on a Windows R installation (file path different, of course) and the script works.
Any idea of why Ubuntu R installation cannot find the odbcConnectExcel function and how I can get this to work?
That functionality is available where Excel is available. In other words: not on Ubuntu.
For reference, from the R Data Import / Export manual (with my highlighting):
4.3.2 Package RODBC
Package RODBC on CRAN provides an
interface to database sources
supporting an ODBC interface. This is
very widely available, and allows the
same R code to access different
database systems. RODBC runs on
Unix/Linux, Windows and Mac OS X, and
almost all database systems provide
support for ODBC. We have tested
Microsoft SQL Server, Access, MySQL,
PostgreSQL, Oracle and IBM DB2 on
Windows and MySQL, Oracle, PostgreSQL
and SQLite on Linux.
ODBC is a client-server system, and we
have happily connected to a DBMS
running on a Unix server from a
Windows client, and vice versa.
On Windows ODBC support is normally
installed, and current versions are
available from
http://www.microsoft.com/data/odbc/ as
part of MDAC. On Unix/Linux you will
need an ODBC Driver Manager such as
unixODBC (http://www.unixODBC.org) or
iOBDC (http://www.iODBC.org: this is
pre-installed in Mac OS X) and an
installed driver for your database
system.
Windows provides drivers not just for
DBMSs but also for Excel (.xls)
spreadsheets, DBase (.dbf) files and
even text files. (The named
applications do not need to be
installed. Which file formats are
supported depends on the the versions
of the drivers.) There are versions
for Excel 2007 and Access 2007 (go to
http://download.microsoft.com, and
search for Office ODBC, which will
lead to AccessDatabaseEngine.exe), the
`2007 Office System Driver'.
I've found RODBC to be a real pain in the Ubuntu. Maybe it's because I don't know the right incantations, but I switched to RJDBC and have had much better luck with it. As discussed here.
As Dirk says, that wont solve your Excel problem. For writing Excel I've had very good luck with the WriteXLS package. In Ubuntu I found it quite easy to set up. I had Perl and many of the packages already installed and had to simply install Text::CSV_XS which I installed with the GUI package manager. The reason I like WriteXLS is the ability to write data frames to different sheets in the Excel file. And now that I look at your question I see that you want to READ Excel files not WRITE them. Hell. WriteXLS doesn't do that. Stick with gdata, like Dirk said in his comments:
gdata on CRAN and you are going to want the read.xls() function:
read.xls("//path//to/excelfile.xls", sheet = 1, verbose=FALSE, pattern, ...,
method=c("csv","tsv","tab"), perl="perl")
you may need to run installXLSXsupport which installs the needed Perl modules.
read.xls expect sheet numbers, not names. The method parameter is simply the intermediate file format. If your data has tabs then don't use tab as the intermediate format. And likewise for commas and csv.