Visualizing Science with ParaView

Joey Bernard — Wed, 29 May 2019 12:00:00 +0000

I'd like to introduce one of the more popular tools used for visualizing data within several scientific disciplines: ParaView. ParaView started as a joint project between Kitware, Inc., and Los Alamos National Laboratory back in 2000. The first public release was version 0.6, which came out in 2002. Since then, ParaView has become one of the most popular visualization packages for visualizing large data sets.

Because it's open source, it should be available in most, if not all, package repository systems. For example, in Debian-based distributions, you should be able to install it with the command:


sudo apt-get install paraview

Starting it the first time should give you an empty workspace, ready for you to get to work.

Figure 1. When you first start ParaView, you'll see a new, empty layout to start your visualization.

Two major parts populate the bulk of the window. The right-hand side is the main display pane where the visualization will appear. The left-hand pane shows the list of objects being visualized, along with their properties. At the top, there is a toolbar of the common functions in ParaView.

To play with ParaView, you'll need some data. If you don't have any data of your own to use, you can grab some data provided as part of the ParaView Tutorial. More documentation and sample scripts are also available there.

Let's assume you're going to use the sample data as you learn how to use ParaView. To load the data, click File→Open, and navigate to where you unpacked the sample data.

While you're here, take a quick look at the list of all of the file types ParaView supports. For example, you can load the data stored in the file can.ex2. You won't see anything displayed right away. In the bottom part of the left-hand side pane, you should see the properties for the newly loaded data file. For now, you can just accept the defaults and click the apply button. You then should see the data visualized in the main pane.

Figure 2. The data in the sample file can.ex2 renders as a half cylinder attached to a rectangle on the end.

Clicking and dragging on the image allows you to rotate the view, so you can see the entire object from various angles.

Go to Full Article

A Good Front End for R

Joey Bernard — Thu, 26 Apr 2018 14:30:00 +0000

by Joey Bernard

R is the de facto statistical package in the Open Source world. It's also quickly becoming the default data-analysis tool in many scientific disciplines.

R's core design includes a central processing engine that runs your code, with a very simple interface to the outside world. This basic interface means it's been easy to build graphical interfaces that wrap the core portion of R, so lots of options exist that you can use as a GUI.

In this article, I look at one of the available GUIs: RStudio. RStudio is a commercial program, with a free community version, available for Linux, Mac OSX and Windows, so your data analysis work should port easily regardless of environment.

For Linux, you can install the main RStudio package from the download page. From there, you can download RPM files for Red Hat-based distributions or DEB files for Debian-based distributions, then use either rpm or dpkg to do the installation.

For example, in Debian-based distributions, use the following to install RStudio:


sudo dpkg -i rstudio-xenial-1.1.423-amd64.deb

It's important to note that RStudio is only the GUI interface. This means you need to install R itself as a separate step. Install the core parts of R with:


sudo apt-get install r-base

There's also a community repository of available packages, called CRAN, that can add huge amounts of functionality to R. You'll want to install at least some of them in order to have some common tools to use:


sudo apt-get install r-recommended

There are equivalent commands for RPM-based distributions too.

At this point, you should have a complete system to do some data analysis.

When you first start RStudio, you'll see a window that looks somewhat like Figure 1.

Figure 1. RStudio creates a new session, including a console interface to R, where you can start your work.

The main pane of the window, on the left-hand side, provides a console interface where you can interact directly with the R session that's running in the back end.

The right-hand side is divided into two sections, where each section has multiple tabs. The default tab in the top section is an environment pane. Here, you'll see all the objects that have been created and exist within the current R session.

The other two tabs provide the history of every command given and a list of any connections to external data sources.

Go to Full Article

Data Analysis

Visualizing Science with ParaView

A Good Front End for R