Home > Blog
We recently updated our "Inference for R Overview" screencast. The new 2-minute-and-5-seconds-long screencast provides a brief overview of how Inference for R can help you fully leverage the power of R in your organization. You can view the overview from our Screencasts page, or by clicking on the image below:
We just released a minor update to our recent Inference for R version 3.5 release. This new update (version 3.5.1) addresses the following items:
- Provides support for R 2.9.1
- Fixes a couple of installation issues users were experiencing
- Copies over any existing Inference license information (to the appropriate new location used by Inference 3.5) during installation
Incidentally, I've improved our R version checking routine in this release such that future releases of R (2.9.2, 2.10.0, etc.) should work fine with Inference without an additional software update. Rather late than never, right? J
You can download the 3.5.1 update from the Download Free Trial page. Make sure you uninstall your existing version of Inference before installing the new one!
We're pleased to announce the release of Inference for R version 3.5. This is a significant release for Inference and includes several major features requested by users:
- Inference in PowerPoint: Embed and run R code in PowerPoint presentations just as you can in Word and Excel.
- Run-in-Place: Execute the R code inside of an Excel, Word, or PowerPoint document and see the code results immediately in the same document – "preview" how your document will execute without having to execute to a separate Results Document.
- Execute as Blog Posting: Take any Inference in Word 2007 document and generate a results document in the Microsoft Word 2007 Blog Post format – you can now easily blog about your R code and results. Office 2007 only.
- Copy and Paste code blocks and expressions between documents.
- The order of expressions can now be changed.
- The built-in Inference R functions are now included in the Object Browser.
- Removed the dependence on the XML and StatDataML R packages – you no longer have to have those packages installed to use Inference for R.
- Inference now installs in a single installation folder selected by the user during installation.
Try it out today by downloading a trial version. If you're a previous Inference for R trial user and your trial has expired, you're in luck – we've reset the trial timer, and you will have another 30-days to evaluate Inference for R version 3.5.
To finish up our series on What's New in Inference for R version 3.5, I'd like to go over a new Inference in Word execution option called Execute to Blog Posting. This feature allows you to take any Inference in Word document and generate a results document in the Microsoft Word 2007 Blog Post format. Before we get started, I should mention that this feature is only available in the Office 2007 version of Inference for R. That's because…well…only Microsoft Word 2007 supports direct blogging. Word 2003 users are unfortunately out of luck…
If you're not familiar with blogging from Word, you can read all out setting up, configuring, and posting blogs using Microsoft Word 2007 here:
Let's take a look at how this actually plays out in Inference in Word. Here's an Inference in Word document that I've created that I would like to post as a blog entry. It's admittedly a contrived example…ok, fine…I modified one of our QuickStart documents. Here it is:
I can then execute this document directly to a Word Blog Posting file from within the Inference Ribbon:
When selected, you will be prompted for where to save the resulting file:
Once the document execution process has completed, you can open your new Blog Post file in Microsoft Word:
You can then edit the document, add your post title, and make any other changes you want. When you're ready to post the blog entry, you use Word's own publish functions:
You'll then be prompted to enter your blog credentials:
…and then Word will upload your blog post (and pictures!) automatically. You can then view the uploaded post in your blog:
Pretty cool, huh? J
One of the most-often heard complaints about Inference is that it's difficult to test an Inference in Word or Excel document as you're writing it, given that the only way to see if it works is to execute the entire document each time to a new file. We've had a number of users who requested some way of previewing the output of an Inference in Word or Excel document without having to execute the document each time to a separate file.
Our solution to this problem is a new feature for Inference for R version 3.5 called Run-in-Place. Run-in-Place allows you to execute the R code inside of an Excel, Word, or PowerPoint document and see the code results immediately in the same document. This allows you to run the code in your document without executing to a separate results document; it in essence provides a "preview" of what your results document will look like. It's also much faster, as it temporarily modifies the open document instead of creating a new one. You can clear the results from the open document at any time, and the existing Execute to a Results Document functionality remains unchanged.
Let's take a look at Run-in-Place in action. We'll start with an Inference in Excel document. When I click "Run Now" from the "Run-In-Place" section of the Inference ribbon:
…Inference extracts and runs the R code from the document:
…and then formats and inserts the results directly into the current document:
All of the results that you would normally see in a Results Document (text, graphics, charts, etc.) are inserted into the current document. And you can clear these results at any time by clicking the "Clear Results" button in the "Run-in-Place" section of the Inference Ribbon. The results are intended to be temporary – a "preview" if you will. For that reason, Run-in-Place results are automatically removed before you save the document. If you want to persist the results, you can execute the workbook to a Results Document using the standard Execute process. Run-in-Place operates significantly faster than Executing to a Results Document, as Run-in-Place acts on the open file and doesn't require saving/opening new files.
Run-in-Place works especially well in Excel, as you can use the input sheet to quickly change input values:
…and then using Run-in-Place, almost immediately see the results:
Run-in-Place works the same way for Inference in Word documents as well. By clicking the "Run Now" button from the Inference Ribbon:
…the R code results are generated and then inserted into the current document:
Inference in PowerPoint (also new for Inference for R version 3.5) presentations work the same way. When we start Run-in-Place on an Inference in PowerPoint presentation:
…all of the text and graphical results generated from the R code are inserted into the current presentation:
Inference for R version 3.5 will be released any day now. Try it out and see if Run-in-Place speeds up your Inference development.
As the release of Inference for R version 3.5 is right around the corner, I thought I'd take a moment to highlight some of the new features. Given the success of Inference in Word and Excel, it was only natural that we would extend those same capabilities to Microsoft PowerPoint:
Inference in PowerPoint allows you to add a Parts Container any PowerPoint presentation:
Once a Parts Container has been added, you can add datasets, R source and data files, and code blocks and expressions directly to the document:
Then, within the PowerPoint slides, you can alias the output from code blocks and expressions to be inserted into the slides when the presentation is executed. You can alias a specific code block or expression using its label plus starting and ending brackets ("<", ">") as placeholders:
When the document is executed, the placeholders are replaced with the actual text and graphical output. When this above document is run-in-place, the slides transform to:
You can also use the placeholders in the notes section of each slide:
When the document is executed, these placeholders are then replaced with the actual text output:
As you can see above, the formatting of the placeholder text will be applied to the output text. You can also output the code from a code block or expression into the slide using the label of the code item plus "_code" in brackets. To make it easier to add the placeholder text, you can right-click on any code block or expression to add the placeholder text for its label (for either text/graphical output or for code output) directly into the slide at the current cursor position:
You can execute the document using either Run-In-Place, which temporarily runs the code and inserts the results into the current document:
…or using by Executing to a Results Document:
…which generates an entirely new PowerPoint presentation as a new file that includes only the formatted presentation without the Parts Container data. You can use this file to email to colleagues, or use to actually show the presentation:
Inference in PowerPoint is brand-new for Inference for R version 3.5. If you have any suggestions or comments, please feel free to email them to me at firstname.lastname@example.org! You'll be able to try out Inference in PowerPoint very shortly; we'll be mailing all existing trial users and newsletter subscribers once the new release is out.
With the release of R version 2.9.0 on 2009-04-17, Inference users have inquired about the best way to upgrade from an earlier R version without losing one's R environment configuration (e.g., installed packages).
Although there are varied ways to upgrade successfully, Inference users have found that the following steps do the trick:
- Uninstall the old R version. From the Windows Start menu, select All Programs > R > Uninstall R <version>. IMPORTANT Note from R FAQ: "Uninstalling R only removes files from the initial installation, not (for example) packages you have installed or updated."
- Install the new version of R. Download R 2.9.0 Windows binary from CRAN and execute the installation in the usual fashion.
- Copy the packages to the new installation. Using Windows Explorer, navigate to the folder corresponding to the libraries of your old R version. (Depending on your OS and R settings, this is likely C:\Program Files\R\<version>\library\ or C:\Users\<UserName>\Documents\R\win-library\<version>\.) Open the library folder, then copy all the packages left behind. Paste them in the library folder of the new installation. If Windows asks whether or not you want to replace an existing folder, select No. After completing the copy, you can delete anything remaining from the old install.
- Update the packages in the upgrade install. In the R Console window, run "update.packages(checkBuilt=TRUE, ask=FALSE)" to update all the packages in the upgrade library folder.
At the useR! 2009 conference, Blue Reference CEO Paul van Eikeren will present a paper illustrating the use of Inference to construct dynamic applications. An abstract of this presentation can be obtained at its linked title below:
Microsoft Office Dynamic Documents as R Applications
The presentation will be on Thursday, July 9, between 14:50 - 15:45 as part of the useR! Focus on Office Integration. If you're going to be attending the useR! 2009 conference, make sure you attend the presentation and say hello!
I'm pleased to announce that we just released a new update (version 3.0.2) to Inference for R. This update includes the following:
- Support for R 2.9
- Support for running Inference for R on 64-bit (x64) versions of Windows XP and Windows Vista
If you're a user of R 2.9 or 64-bit Windows, you should download a new version today. Incidently, if your Inference trial has expired and you would like to try this updated version, please feel free to contact me at email@example.com for an extended trial license key.
On a related note, we're still on-track for a June release for the next major release of Inference for R, which will include Inference in PowerPoint and the much-requested Run-In-Place functionality.
||© 2009 Blue Reference, Inc.
All rights reserved. | 3052 NW Merchant Way, Suite 100, Bend, Oregon 97701 | 541-316-2343 |