Tag Archive: R


ggplot2 blog

I see Hadley Wickham has launched a new ggplot2 blog.

I got to meet Hadley for a day long tutorial session and later share dinner. I am truly impressed by what he has delivered and by how he continues to foster its growth.

If you are not familiar with ggplot2, it is a graphical package for the R project.

Personally, ggplot2 was my gateway into the land of R and I still very much use it almost daily. For me, it provided the building-block type of functionality that allowed me to gradually learn. Its strength for me was how readily I could re-use each building-block that I learned.

All in all, I look forward to seeing ggplot2 grow and I’m thankful for Hadley’s contributions.

R GoogleVis Rocks!

I tried out the GoogleVis package for R today as a nice distraction. It was indeed a satisfying distraction.

It’s probably old news for many but it was the first time I had played with it.

I really only explored the motion chart. The others didn’t seem quite as appealing.

My only complaint was that it did not support timestamp based time series. I believe that is a limitation of the Google implementation though.

Would be curious to further explore implementing this via R-Apache but I didn’t get this far.

Below you can see the examples I created. The first is the most fun, moving scatter plot with configurable axis, colors, and sizes.

The second is a moving bar chart that has some configurables and lastly is a line chart.

The R code to create this is very simple:

g = gvisMotionChart(dbqldata,idvar="WorkLoadName",timevar="TheDate",options=list(width=700, height=600))

More info at these links:

For an assortment of reasons I decided I needed to change up my RStudio Server . In a previous article I alluded to some challenge of building from source on CentOS but there still were some lingering problems. So I went to a packaged releases of R and RStudio Server.

From R perspective, things were find. For RStudio Server, less so. Install was fine. However once I logged in it would just hang. Of course logging is not very helpful and very sparse with RStudio and I didn’t mess around with it very much after initial problems.

The other day I got the .95 version and decided to give it a try. I installed and had same problem. This time I noticed a reference of libodbc in /var/log/messages along with rsession’s segfault error message. I did an ldd on rsession and there was no reference. I thought where could it be coming from?

I deleted all the rstudio temporary and user files but still hit problem. Then I remembered my .Rprofile had a “library(RODBC)” entry. I took it out and IT WORKED! I added it back and IT WORKED! So I can’t really explain why it was failing but I can say it works now.

Oh, and .95 has SVN and GIT integration built in which is nice. Supports a new “project” perspective which is nice.

Rstudio and R on CentOS — aaagh

Could this be more annoying?

Granted I didn’t use the RPM’s because I didn’t want X11 but jeez..

And what’s with more and more projects using cmake? How many more build tools do we need?

The tricks were to execute the get dependencies section in rstudio before doing cmake stuff and then change default port to something not already registered in services and lastly add the firewall rule. Perhaps the most frustrating aspect was the lack of any logging going while things aren’t working as desired.

So there’s my rant as well as tips to getting it to work.

Follow

Get every new post delivered to your Inbox.

Join 25 other followers