LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

support vector macine

hello

 

I need to know how to emplement support vector machine in lab view for classification of two types of signals. how to train it . how to test it .(any knowledge will be helpful)

 

 

thank you so much .

0 Kudos
Message 1 of 6
(3,364 Views)

we need more detail,   less font size(would be nice) and if English is not your first language, go ahead and post in your native tounge.   We can google translate without straining ourselves

 

 

Welcome to the forums


"Should be" isn't "Is" -Jay
0 Kudos
Message 2 of 6
(3,352 Views)
What is the signa type? And as bohrer said you can post in your own language..
0 Kudos
Message 3 of 6
(3,342 Views)

There are plenty of handy VIs that can be used to hammer down some of the characteristics of the signals being read in found under the Signal Processing>>Waveform Measurements palette such as:

 

Pulse Measurement.vi

Transition Measurement.vi

Amplitude and Level.vi

Cycle Average & RMS.vi

etc.

 

I suppose the best way to categorize your 2 different types of signals would be to use these to create some sort of a "Finger print" for your SVM to use to learn what its 2 signal categories are and compare to new signals coming in.

 

I'm going to have to back up the other two responders though in saying I need more info to go from here.

 

Logan H

National Instruments
Applications Engineer
0 Kudos
Message 4 of 6
(3,322 Views)
sorry, I did not mean to shout . I just wanted the the font to be easier to be read. I am acquiring EEG signal using serial port. I have an array of boleen LEDs in which every LED has a letter written on it . These LEDs light up in a certain sequence . The user should concentrate on one letter written on one LED to write it with his thoughts . lighting the intended letter is a visual stimuli that generates an unordinary signal called p3 . This signal appears after 300 ms of the occurence of the stimuli .So, I need to extract the segment between 150 and 600 ms after every LED lights up .Then I will use these segments to train a machine learning technique called "support vector machine " to distiguish the p3 from the ordinarry signals . and then perform some way to identify the intended LED ,and so the corresponding letter. I hope that I could clear what I wanted this time. Any ideas ? Thank you .
0 Kudos
Message 5 of 6
(3,306 Views)

You've posted this in three different threads. This one here, this one, and this one.

 

Please pick one thread so all of the information is in one place. 

0 Kudos
Message 6 of 6
(3,298 Views)