Category Archives: Eigenvector News
EigenU 2011 Filling Up!
Mar 31, 2011
The Sixth Edition of Eigenvector University, EigenU 2011, is filling up. Many of our most popular classes are more than half full. This includes Linear Algebra for Chemometricians, MATLAB for Chemometricians, Chemometrics I: PCA, Chemometrics II: Regression and PLS, Variable Selection, and Common Chemometric Mistakes (and how to avoid them).
This year EigenU runs from Sunday, May 15 through Friday, May 20. As before, it will be held at the nation’s premier city athletic club, the Washington Athletic Club, in the heart of Seattle.
Early discount registration ends soon. To receive the early discount rates, registration must be received with payment by April 15.
See you in May!
BMW
Improved Model_Exporter Coming Soon
Mar 29, 2011
EVRI’s Model_Exporter allows users of PLS_Toolbox and Solo to export their Principal Component Analysis (PCA), Partial Least Squares (PLS) Regression and other models in a variety of formats suitable for use in both MATLAB and other environments (e.g. LabSpec, GNU Octave, Symbion, LabVIEW, Java, C++/C#). A new, improved version of Model_Exporter will be released in the coming weeks. The new version will allow models to be more easily applied on platforms with less computer power, such as on hand-held devices.
The new Model_Exporter will include a number of improvements including a much less memory-intensive encoding of Savitzky-Golay derivatives and smoothing and similar improvements in handling of excluded variables. The new version will also support additional preprocessing methods which were released in versions 6.0 and 6.2 of PLS_Toolbox and Solo.
The other big news is that we will also be bundling with Model_Exporter a freely-distributable .NET class to do predictions from exported models. As-is, this class can be integrated into any .NET application (C++/C#/VB) from Microsoft Visual Studio 2005 or later. It can be used and redistributed without any per-use charge (exactly as the exported models are licensed) and we’d be happy to work with you on integrating this class into your end application. If you have interest in this approach, just let our Chief of Technology Development, Jeremy Shaver, or me know.
BMW
New Versions Released
Mar 23, 2011
New versions of our most popular software products were released earlier this month. Our MATLAB-based, full-featured chemometrics package, PLS_Toolbox, is now in version 6.2. The stand-alone version of PLS_Toolbox, Solo, and Solo+MIA for multivariate and hyperspectral image analysis also move to version 6.2. MIA_Toolbox is now in version 2.5.
The functionality of Solo+MIA and MIA_Toolbox has been expanded greatly with the integration of ImageJ, the image processing application developed by NIH. New tools include particle counting with size and shape analysis, interactive data navigator for drilling into an image, magnify, cross-section and image alignment. A number of new data importers have also been added (RAW, BIF and ENVI).
For a tour of the new features, highlighting the new MIA tools, please watch this short video. For a complete list of new features, please see the PLS_Toolbox/Solo 6.2 release notes, and the MIA_Toolbox/Solo+MIA release notes.
Users with current maintenance contracts can download the new versions from their account. Free 30-day demos are also available for download.
We trust that the new tools will make you even more productive!
BMW
Happy Anniversary, Jeremy!
Feb 15, 2011
Time flies! Today marks the 10th anniversary of Jeremy Shaver joining Eigenvector Research. There are so many nice things I could say about Jeremy that I’m not sure where to start.
When Jeremy joined us in 2001, It didn’t take long for us to figure out that he was more than merely multi-talented. He rapidly took over as lead developer of our software products. As Chief of Technology Development, Jeremy, more than anyone else at EVRI, has been the driving force behind the updates (PLS_Toolbox versions 3.0 through 6.0) and many of the new products (MIA_Toolbox, Solo, Model_Exporter, etc.) we’ve released. There is a lot of his vision in the way our software works, looks and feels.
There have been lots of times when one of us said, “Wouldn’t it be great if our software did _____?” And then about 2 hours later Jeremy sends us some code to try out, and says, “You mean like this?” And we just laugh in amazement, “Yeah, like that!”
Jeremy also contributes greatly on consulting projects and is an enthusiastic short course instructor. He always leads our PowerUser Tips & Tricks session at EigenU because nobody know how to get more out of the software than he does.
Besides his chemometric skills, Jeremy is a pilot, a great guitarist (witness this jam session with Rasmus and Randy at EigenU 2010), avid outdoorsman, and devoted husband and dad. And we like to drink beer and hang out with him too!
Congratulations Jeremy. We hope you’ve enjoyed it as much as we have!
BMW
EigenU 2011 Registration Now Open
Jan 25, 2011
The Sixth edition of Eigenvector University will be held May 15-20, 2011. Once again, we’ll be at the fabulous Washington Athletic Club in the heart of downtown Seattle.
We’ve added two new classes this year, “SVMs, LWR and other Non-linear Methods for Calibration and Classification” and “Design of Experiments.” The non-linear methods class will focus on Support Vector Machines for regression and classification, which were added to PLS_Toolbox/Solo version 5.8 (February 2010) along with an interface to Locally Weighted Regression. We’ve found these methods to be quite useful in a number of situations, as have our users. The DOE course will focus on practical aspects experimental design, including designing data sets for multivariate calibration.
I sometimes say that the secret to Eigenvector Research is data preprocessing, i.e. what you do to the data before it hits a PLS or other multivariate model. Thus, “Advanced Preprocessing” has been expanded to a full day for EigenU 2011. We’ll cover many methods for eliminating extraneous variance, including the “decluttering” methods (Generalized Least Squares Weighting, External Parameter Orthogonalization, etc.) we’ve highlighted recently. “Multivariate Curve Resolution” has also be expanded to a full day in order to better cover the use of constraints and contrast control.
EigenU 2011 will also include three evening events, including Tuesday night’s “PLS_Toolbox/Solo User Poster Session,” (with iPod Nanos for the best two posters), Wednesday night’s “PLS_Toolbox/Solo PowerUser Tips & Tricks,” and the Thursday evening dinner event.
You can register for EigenU through your user account. For early discount registration, payment must be received by April 15. Questions? E-mail me.
See you at EigenU!
BMW
Sijmen de Jong
Nov 8, 2010
Sijmen de Jong passed away on October 30, 2010.
I first heard of Sijmen not long after finishing my dissertation. It was late 1992 or so, and I was working at Battelle, Pacific Northwest National Laboratory (PNNL). Sijmen had gotten a copy of an early version of PLS_Toolbox which was available on the internet via FTP. I received a letter from Sijmen with a number of “suggestions” as to how the toolbox might be improved. As I recall, the letter ran at least 3 pages, and included two floppy disks (5.25 inch!) filled with MATLAB .m files.
I was initially taken aback by the letter, and recall thinking, “Who is this guy?” But it didn’t take me long to figure out that, wow!, there was a lot of good stuff there. Several of the routines were incorporated into the next release of PLS_Toolbox, and modifications were made to several others. I’m sure there is still code of his in the toolbox today!
Sijmen was especially interested in the work I’d done with Larry Ricker on Continuum Regression (CR) [1], as he had just published on another continuously adjustable technique, Principal Covariates Regression (PCovR) [2]. He was also working on his SIMPLS [3] algorithm for PLS around that time. Sijmen suggested that the SIMPLS algorithm might be extended to CR. This started a collaboration which would eventually produce the paper “Canonical partial least squares and continuum power regression” in 2001 [4]. I still think that this is the best paper I’ve ever had the pleasure of being associated with. It won the Unilever R&D Vlaardingen “Author Award,” and I still display the certificate proudly on my office wall.
While CR is primarily of academic interest, SIMPLS has become perhaps the most widely used PLS regression algorithm. The reasons for this are evident from several of my recent posts concerning accuracy and speed of various algorithms. If you have done a PLS regression in PLS_Toolbox or Solo, you have benefited from Sijmen’s work!
Sijmen was the kind of smart that I always wanted to be. He seemed to see clearly through the complex math, understand how methods are related, and see how a small “trick” might greatly simplify a problem. As another colleague put it, “A very clever guy!” Beyond that, he was easy to work with and fun to be around. I regret that I got to be around him socially only a few times.
Rest in peace, Sijmen. You will be missed!
BMW
[1] B.M. Wise and N.L. Ricker, “Identification of Finite Impulse Response Models with Continuum Regression,” J. Chemometrics, 7(1), pps. 1-14, 1993.
[2] S. de Jong, H.A.L. Kiers, “Principal Covariates Regression: Part 1, Theory,” Chemo. and Intell. Lab. Sys., Vol 14, pps. 155-164, 1992.
[3] S. de Jong, “SIMPLS: an alternative approach to partial least squares regression,” Chemo. and Intell. Lab. Sys., Vol. 18, pps. 251-263, 1993.
[4] S. de Jong, B.M. Wise and N.L. Ricker, “Canonical partial least squares and continuum power regression,” J. Chemometrics, 15(1), pps. 85-100, 2001.
Advanced Features in Leuven, Belgium
Oct 4, 2010
Following a successful premier of our new course, Using the Advanced Features of PLS_Toolbox/Solo 6.0, we’re pleased to announce that we will be repeating the course in Leuven, Belgium on November 22, 2010. The course is being arranged in conjunction with CQ Consultancy.
And, yes! It was a successful premier! We had 28 students at the initial offering on the University of Barcelona campus. Many, many, thanks to Romà Tauler and Anna de Juan for hosting the event, and of course many thanks to the attendees for coming. We used a beta of Solo 6.0 for the course, and I’m happy to say it ran quite smoothly. It should be ready for release by mid-October.
BMW
Mediterranean Tour
Sep 17, 2010
I head off tomorrow morning (way too early!) on a “tour” of the Mediterranean. I’ll be attending The First African-European Conference on Chemometrics, aka Afrodata, in Rabat, Morocco. From there, I’ll go to CMA4CH–Application of MVA and Chemometrics to Cultural Heritage and Environment in Sicily. Then it’s on to Barcelona to teach Using the Advanced Features of PLS_Toolbox.
I’m looking forward to getting out of the office and spending some time with my chemometric colleagues. When I see something of chemometric interest during my travels, I’ll try to get it posted here. I hope to see many of you in the coming weeks!
BMW
Advanced Features in Barcelona
Sep 1, 2010
I’m pleased to announce that I’ll be doing a one day short course, Using the Advanced Features of PLS_Toolbox, on the University of Barcelona campus, on October 1, 2010 (one month from today!). The course will show how to use many of the powerful, but often underutilized, tools in PLS_Toolbox. It will also feature some new tools from the upcoming PLS_Toolbox 6.0, to be released this fall.
Normally, our courses focus on developing an understanding of chemometric methods such as PCA and PLS. This course is something of a departure in that it focuses on getting the most out of the software. This will give us a chance to show how to access many of the advanced features and methods implemented in PLS_Toolbox.
You can get complete information, including registration info and a course outline, on the course description page. If you have additional questions or suggestions for demos you’d like to see, contact me.
See you in Barcelona!
BMW
10,000 Commits
Jul 30, 2010
The EVRI software developers surpassed a landmark when our Donal O’Sullivan made the 10,000th “commit” to our software repository. Donal was working on some improvements to our Support Vector Machine (SVM) routines in PLS_Toolbox and submitted the changes yesterday afternoon.
As noted by our Chief of Technology Development, Jeremy Shaver, “This is a trivial landmark in some ways (it is just a number, like when your car rolls over 10,000 miles) but it also indicates just how active our development is. It all started with a revision by Scott in March, 2004. In six years time, we’ve committed thousands upon thousands of lines of code and many megabytes of files. That’s an average of 1666 commits/year or 4.6 commits/day although nearly 2000 of those were in the last year.”
The level of activity in our software repository truly demonstrates how our product development continues to accelerate. Thanks go to our users for driving, and funding, the advancements. And of course, to our developers, thanks for all your efforts, guys!
BMW
Eigenvector Welcomes Randy Bishop
Jul 26, 2010
All of us at EVRI would like to issue a warm, (albeit belated), welcome to Randy Bishop. Randy joined our staff in March, 2010.
We started running into Dr. Bishop about 10 years ago at FACSS meetings where he often taught experimental design. In those days he was with GE Plastics and heavily involved with Six Sigma. Since then he has worked in Process Analytical Technology (PAT) with GlaxoSmithKline and Wyeth (now Pfizer).
Randy has a wealth of experience with a broad variety of analytical methods, especially Raman spectroscopy, but also many other types of spectroscopy, spectrometry, and chromatography. He has used multivariate methods extensively and has been a leader in promoting their use among his colleagues. In fact, we’d worked with Randy to implement specific chemometric methods to make it easier for his co-workers to use them.
Beyond that, he’s fun to work with, a great guitarist, and we just love listening to his East Tennessee drawl. We look forward to working with Randy on consulting projects, our short courses (look for some new DOE offerings soon) and software development.
Welcome aboard, Randy!
BMW
New Website Up and Running
Jul 21, 2010
We’re pleased to announce that the new Eigenvector website is up! We’ve been working on it for several months. You may have noticed we haven’t updated the current site during that time, but it is all up-to-date now!
I developed Eigenvector’s first website in the fall of 1996. There were just over 2 million registered domain names in 1996 with .net, .org and .com extensions. Now there are well over 100 million. The original Eigenvector website ran off a server in our house in Manson, WA, which was connected to the internet via frame-relay. It even had a webcam that looked out my window.
Visitors should find the new site more streamlined, consistent and easier to navigate. We’ve also designed it to make it easier for us to keep updated. Our old site, which had grown slowly since it’s last major revamp in about 1999, had become pretty fractured and unwieldy. We hope that we can serve you better with our new site!
BMW
Eigenvector University 2010 Best Posters
May 26, 2010
The EigenU 2010 Poster Session was held Tuesday evening, May 18. This year 7 users contributed posters for the contest, and the Eigenvectorians chipped in another 5, so there was plenty to discuss. We all enjoyed beer, wine and hors d’oeuvres as we discussed the finer points of aligning chromatograms and calculating PARAFAC models, among other things!
This year’s posters were judged by Bruce Kowalski, guest instructor at this year’s EigenU, and Brian Rohrback, President of Infometrix. Bruce and Brian carefully reviewed the submitted posters (not counting the ones from the Eigenvectorians, of course). Thanks to the judges, especially Brian, who stopped by just for this event!
![EigenU_2010_Poster_Winners Barry Wise, Jamin Hoggard, Bruce Kowalski, Cagri Ozcaglar and Brian Rohrback](http://www.eigenvector.com/images/EigenU_2010_Poster_Winners.jpg)
Barry Wise, Jamin Hoggard, Bruce Kowalski, Cagri Ozcaglar and Brian Rohrback
The winners, pictured above, were Cagri Ozcaglar of Rensselaer Polytechnic Institute for Examining Sublineage Structure of Mycobacterium Tuberculosis Complex Strains with Multiway Modeling, and Jamin Hoggard of the University of Washington for Extended Nontarget PARAFAC Applications to GC×GC–TOF-MS Data. Jamin and Cagri accepted iPod nanos with the inscription Eigenvector University 2010 Best Poster for their efforts.
Congratulations, Cagri and Jamin!
BMW
Biggest Chemometrics Learning Event Ever?
May 26, 2010
Eigenvector University 2010 finished up on Friday afternoon, May 21. What a week! Six days, 17 courses, 10 instructors and 45 students. I’d venture a guess that this was the biggest chemometrics learning event ever. If more chemometrics students and instructors have been put together for more hours than in Seattle last week, I’m not aware of it.
Thanks to all that came for making it such a great event. It was a very accomplished audience, and the discussions were great, both in class and over beers. The group fielded lots of good questions, observations and related much useful experience.
We’re already looking forward to doing it next year and have been busy this week incorporating student feedback into our courses and software. The Sixth Edition of EigenU is tentatively scheduled for May 15-20, 2011. See you there!
BMW
Ready for EigenU
May 14, 2010
Eigenvector University 2010 starts in just 2 days. We’re busy doing the final tune-ups on our course notes, making final catering arrangements, etc. I hope that all our guests are ready to get out of their offices for a while and spend some time “sharpening the saw.”
This year’s EigenU will be the biggest ever, with over 45 attendees (not everybody is there every day) plus 10 Eigenvectorians (Neal Gallagher, Jeremy Shaver, Bob Roginski, Scott Koch, Donal O’Sullivan, Randy Bishop, Willem Windig, Bruce Kowalski, Rasmus Bro and myself). So while things might be a bit crowded at times, we’ll have plenty of staff on hand to help with questions whether they’re on the practical aspects of the software or philosophical aspects of multivariate modeling.
Our first evening event is the MATLAB/PLS_Toolbox User Poster session on Tuesday, 5:30-7:30. I’m pleased to announce that the poster contest judges will be Bruce Kowalski, and Brian Rohrback, President of Infometrix, makers of the Pirouette chemometrics package. The two best posters win iPod nanos. We’ll let you know who won!
BMW
Robust Methods
May 12, 2010
This year we are presenting “Introduction to Robust Methods” at Eigenvector University. I’ve been working madly preparing a set of course notes. And I must say that it has been pretty interesting. I’ve had a chance to try the robust versions of PCA, PCR and PLS on many of the data sets we’ve used for teaching and demoing software, and I’ve been generally pleased with the results. Upon review of my course notes, our Donal O’Sullivan asked why we don’t use the robust versions of these methods all the time. I think that is a legitimate question!
In a nutshell, robust methods work by finding the subset of samples in the data that are most consistent. Typically this involves use of the Minimum Covariance Determinant (MCD) method, which finds the samples that have a covariance with the smallest determinant, which is a measure of the volume occupied by the data. The user specifies the fraction, h, to include, and the algorithm searches out the optimal set. The parameter h is between 0.5 and 1, and a good general default is 0.75. With h = 0.75 the model can resist up to 25% arbitrarily bad samples without going completely astray. After finding the h subset, the methods then look to see what remaining samples fall within the statistical bounds of the model and re-include them. Any remaining samples are considered outliers.
The main advantage of robust methods is that they automate the process of finding outliers. This is especially convenient when the data sets have many samples and a substantial fraction of bad data. How many times have you removed an obvious outlier from a data set only to find another outlier that wasn’t obvious until the first one is gone? This problem, known as masking, is virtually eliminated with robust methods. Swamping, when normal samples appear as outliers due to the confidence limits being stretched by the true outliers, is also mitigated.
So am I ready to set my default algorithm preferences to “robust?” Well, not quite. There is some chance that useful samples, sometimes required for building the model over a wide range of the data, will be thrown out. But I think I’ll at least review the robust results now each time I make a model on any medium or large data set, and consider why the robust method identifies them as outliers.
FYI, for those of you using PLS_Toolbox or Solo, you can access the robust option in PCA, PCR and PLS from the analysis window by choosing Edit/Options/Method Options.
Finally, I should note that the robust methods in our products are there due to a collaboration with Mia Hubert and her Robust Statistics Group at Katholieke Universiteit Leuven, and in particular, Sabine Verboven. They have been involved with the development of LIBRA, A MATLAB LIBrary for Robust Analysis. Our products rely on LIBRA for the robust “engines.” Sabine spent considerable time with us helping us integrate LIBRA into our software. Many thanks for that!
BMW
Best Poster iPods Ordered
Apr 28, 2010
As in past years, this year’s Eigenvector University includes a poster session where users of MATLAB, PLS_Toolbox and Solo can showcase their work in the field of chemometrics. It is also a chance to discuss unsolved problems and future directions, over a beer, no less.
This year’s crop of posters will be judged by Bruce Kowalski, co-founder of the field of chemometrics. For their efforts, the top two poster presenters will receive Apple iPod nanos! This year I’ve ordered the 16GB models that record and display video and include an FM tuner with Live Pause. These are spiffy, for sure. We’ll have one in blue (just like the EVRI logo!) and one in orange (our new highlight color, new website coming soon!). Both are engraved “Eigenvector University 2010, Best Poster.”
There is still time to enter the poster contest. Just send your abstract, describing your chemometric achievements and how you used MATLAB and/or our products, to me, bmw@eigenvector.com. Then be ready to present your poster at the Washington Athletic Club in downtown Seattle, at 5:30pm on Tuesday, May 18. The poster session is free, no need to register for EigenU classes.
You could win!
BMW
EigenU 2010 on-track to be biggest ever
Apr 18, 2010
The Fifth Edition of Eigenvector University is set for May 16-21, 2010. Once again we’ll be at the beautiful Washington Athletic Club in Seattle. This year we’ll be joined by Bruce Kowalski, co-founder (with Svante Wold) of the field of Chemometrics. Rasmus Bro will also be there, along with the entire Eigenvector staff.
Registrations are on-track to make this the biggest EigenU ever. All of the 17 classes on the schedule are a go! We’re also looking forward to Tuesday evening’s Poster Session (with iPod nano prizes), Wednesday evening’s PowerUser Tips & Tricks session, and Thursday evening’s Workshop Dinner. It is going to be a busy week!
BMW
Another EAS Meeting
Nov 15, 2009
Hard to believe that a year has passed, but I’m once again back at the Eastern Analytical Symposium, in Somerset NJ. EAS 2009 started today (Sunday, Nov. 15) with short courses, and will continue through Thursday (Nov. 19).
I’m teaching today with Don Dahlberg. Once again we’re presenting “Chemometrics without Equations” to a new group of chemometric neophytes. It’s a good class this year, a dozen registrants, all with real-world data problems they’d like to solve.
The trade show part of EAS starts tomorrow. I’m happy to report that our booth arrived safely this year. I’m excited to show off the new versions of our software, including version 5.5 of PLS_Toolbox, Solo and Solo+MIA, and version 2.0 of MIA_Toolbox. If you are in the neighborhood, please drop by booth #329 for a demo. Mention this blog and I’ll give you a free 2G USB drive! (Limited to first 50 requests.)
I’m also looking forward to the session for the Achievements in Chemometrics Award, which Eigenvector sponsors. This year Romà Tauler is the award recipient. He is being recognized for his work with curve resolution methods, and it is certainly well-deserved!
I’ll check back with another report on EAS 2009 later this week.
BMW
Chemometric “how to” videos on-line
Nov 12, 2009
If a picture is worth a thousand words, what’s a video worth?
Here at EVRI we’ve started developing a series of short videos that show “how to” do various chemometric tasks with our software packages, including our new PLS_Toolbox 5.5 and Solo 5.5. Some of the presentations are pretty short and specific, but others are a little longer (10-15 minutes) and are a blend of teaching a bit about the method being used while showing how to do it in the software.
An example of the latter is “PCA on Wine Data” which shows how to build a Principal Components Analysis model on a small data set concerning the drinking habits, health and longevity of the population of 10 countries. Another movie, “PLS on Tablet Data” demonstrates building a Partial Least Squares calibration for a NIR spectrometer to predict assay values in pharmaceutical tablets.
While I was at it, I just couldn’t help producing a narrated version of our “Eigenvector Company Profile.”
We plan to have many more of these instructional videos that cover aspects of chemometrics and our software from basic to advanced. We hope you find each of them worth at least 1000 words!
BMW