Anders Björk

QUOTE (WAtoMA @ Jan 27 2009, 01:53 AM) Have you seen the tabcontrol i Labview, use a state machine for handling the different cases? Using that you dont need to have three separate windows. My suggestion is put the instrument read/control in parallell what you need to present. A starting point is getting the read of raw values to work.

Hi I have an activex variant that gives this type of data (which I have get by a calling IProjectData): A pointer to IProjectData in which the data matrix contains: First dimension of the variant = number of observations in the model Second dimension of the variant = number of components in the model+1 (the 0 component). I have tried different ways to get the data out from the variant, making an array of cluster among other tested approaches. Have any of you some suggestions what to do? Some of the objects that I can get through IProjectdata are matrixes (with numerics) no problem with them only with those mentioned above.

QUOTE (Ton @ Jan 17 2009, 11:30 PM) Do you have control software from the manufacture, using portmon seems to be a good solution to get the control codes as well as settings for the serialport! http://technet.microsoft.com/en-us/sysinte...s/bb896644.aspx

QUOTE (freemason @ Jan 15 2009, 09:19 PM) http://lavag.org/old_files/monthly_01_2009/post-5713-1232101435.jpg' target="_blank"> Something like this, maybe I guess it could be made simpler and better than this but this seem to work.

QUOTE (freemason @ Jan 15 2009, 09:19 PM) http://lavag.org/old_files/monthly_01_2009/post-5713-1232101435.jpg' target="_blank"> Something like this, maybe I guess it could be made simpler and better than this but this seem to work.

QUOTE (torekp @ Jan 14 2009, 01:38 PM) I would never set the stop critera as a tolerance, you easily overfit the model. Instead test by crossvalidation or test set the reasonable numbers of PLS-components. Another criterar for max PLS-comp is how many driving forces generate you data or affect the class divisions.

What will you use your data for later?

A exponential filter could possibly be useful http://www.statistics.com/resources/glossary/e/expfilt.php You will have experiment with the alpha value to get a good filtering effect.

QUOTE (hanady @ Dec 29 2008, 02:28 PM) Show us your code! Without that we can have no clues what is faulty! More information about your hardware is also needed. BR Anders

QUOTE (torekp @ Dec 16 2008, 05:58 PM) Great! I would like to test it, if you get it working.

On this page there seems to be some matlabcode to do MDA http://www.cs.tut.fi/sgn/arg/SGN-2556/ Here you have a more easy read discription I think, from looking at it fast http://gupea.ub.gu.se/dspace/bitstream/207...l_Gusev_IFE.pdf I would do Partial Least Squares Discriminant analysis, however that requires code to PLS.

QUOTE (Minh Pham @ Nov 25 2008, 02:13 AM) Thank you very much Minh Pham this helps a great deal!

Hi all I am looking for any has any ideas how set all (or selected ones) graphs (in frontpanel) X-axis minimum and mixumvalues at the same time? For instance I want to set the time span to 5 days, 24h or 8h. I have look at getting all control reference from a VI but not getting any further than that.

QUOTE (DanielChile @ Nov 21 2008, 11:11 PM) If you would define all PCs admin password, and you would have unattend installation package this would not be so hard for NI make.

Look at this http://www.activexperts.com/activmonitor/w...ts/disk/drives/ I think have to VB-script or the COM object directly.

QUOTE (normandinf @ Nov 13 2008, 12:58 AM) Well for microphones you have one base amplitude and accelerometers on other. So log of a ratio Ayour/Abase. Look it up on wikipedia or some other place. For the A-weighting part there are a formula for that on wikipedia as well if I remember correctly.

Could you give an example how much less i precent it gets? I think it has to do with the scaling the y-axis of the powerspectra, you could look it up in DSP book.

QUOTE (pabhon002426 @ Nov 7 2008, 08:24 PM) Break down the problem, write code (VI) for the voltmeter first, test your code, improve it, ok?, then do the VI for the signal generator etc... Learn.. If you have specific problems ask here. If you have low labview knowledge read some good bok or look at labview for dummies atleast if it is still on the internet...

QUOTE (torekp @ Oct 22 2008, 02:58 PM) Adding the vectors to a matrix common is to enable sorting of the eigenvecors in order from large to small, which i common in PCA. QUOTE (torekp @ Oct 22 2008, 03:59 PM) And here is my attempt to follow this recipe (thanks, Los Alamos!) using SVD. I changed from normalizing my data matrix to merely centering it, but I changed my earlier VI similarly and the results agree. Hooray! Does this mean I actually did this right? According to the website, Vector S^2 is proportional to the variances of the principal components, so I'm taking that as a measure of how important each Component is. A PCA-modell is usally like this X=TP'+E Where T=S*(What you call scores now) multiplying Scorevector1*S1, Scorevector2*S and so on. Now youre PCA-model is factored in two instead of three variables. Sorry for the late answer, had about too much work for a time.

Are the images supposed to be similar all the time? I mean is it a limited set of images? Checking sqrt(sum((blue-blueavg)^2+(green-greenavg)^2+(red-redavg)^2))/2) could be a way.

How about using medianfilter instead of averaging?

QUOTE (torekp @ Oct 20 2008, 02:39 PM) If youre matrices is complete it can be done by SVD plus four of five other VIs.

QUOTE (ivan00 @ Oct 20 2008, 02:07 PM)

QUOTE (torekp @ Oct 16 2008, 07:55 PM) Hello Your datatable is 10k observations and 1k variables? Then I would build a PCA-model and do clustering on the PCA-scores, to get an reasonable dimension for each observation. When get new data your could use the PCA-loadings to calculate new PCA-scores.

Is the software installing an usb-serieport driver? In that case it would be fairly easy to look at the communication by portmonitor program, like portmon från microsoft.