<div dir="ltr">Dear Karlo,<div><br></div><div style>I saw it when my data were very dirty or rank-reduced.</div><div><br></div><div style>Run following codes and tell me what you see.<br></div><div style>1. figure; bar(std(EEG.data(:,:),0,2));</div>
<div style>Use data reader tool (find it from matlab figure icons). If you see any bar standing out, read its channel number (let's say it is channel N) and run next one.</div><div style>2. figure; plot(EEG.data(N,:))</div>
<div style>Also run</div><div style>3. figure; bar(mean(EEG.data(:,:)));</div><div style><br></div><div style>Makoto</div><div class="gmail_extra"><br><div class="gmail_quote">2013/5/6 karlo gonzales <span dir="ltr"><<a href="mailto:thats_karlo@yahoo.com" target="_blank">thats_karlo@yahoo.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div style="font-size:10pt;font-family:arial,helvetica,sans-serif"><div style="font-family:arial,helvetica,sans-serif;font-size:10pt">
Dear Friends,</div><div style="font-family:arial,helvetica,sans-serif;font-size:10pt"><br></div><div style="background-color:transparent"><span><span><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">I faced a problem with running ICA, that required your expertise and guides. </span></span></span></span><span style="background-color:transparent;font-family:arial,helvetica,sans-serif;font-size:13px"><span><span>I am simply using: </span></span></span><span style="background-color:transparent"><span style="font-family:arial,helvetica,sans-serif;font-size:13px;font-weight:bold">EEG = pop_runica(EEG,'icatype','runica'); </span><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">but it seems because of the nature of my data (rank , Gaussian, or ....), ICA algorithm needs to lower learning rate frequently after few steps (please see the below). size of my data is >> size(EEG.data) >> ans= 62 4820 159. </span></span></span></div>
<div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif"><span style="background-color:transparent"><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px"><br>
</span></span></span></div><div style="background-color:transparent"><span style="background-color:transparent"><span><span><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">- how do you check rank of your EEG data? (rank(EEG.data) gives error for 3d
matrix )</span></span></span></span></span></div><div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif">- what could be wrong here or what i need to check before running ICA?</div>
<div style="background-color:transparent"><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">- May i ask you that how do you choose your ICA options (e.g., 'pca' , 'extended1' or ...)? </span></span></div>
<div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif"><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px"><br></span></span></div><div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif">
<span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">Many thanks,</span></span></div><div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif">
<span><span style="font-family:arial,helvetica,sans-serif;font-size:13px">Karlo </span></span></div>
<div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif"><span style="background-color:transparent"><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px"><br>
</span></span></span></div><div style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,sans-serif"><span style="background-color:transparent"><span><span style="font-family:arial,helvetica,sans-serif;font-size:13px"> </span></span></span></div>
<div style="background-color:transparent"><span style="font-family:arial,helvetica,sans-serif;font-size:13px"> </span></div><div style="background-color:transparent"><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Input data size [62,766380] = 62 channels, 766380 frames/nFinding 62 ICA components using logistic ICA.</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Decomposing 199 frames per ICA weight ((3844)^2 = 766380 weights, Initial learning rate will be 0.001, block size 68.</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Learning rate will be multiplied by 0.9 whenever angledelta >= 60 deg.</font></div><div style="background-color:transparent">
<font face="arial, helvetica, sans-serif" style="font-style:italic">More than 32 channels: default stopping weight change 1E-7</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Training will end when wchange < 1e-007 or after 512 steps.</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Online bias adjustment will be used.</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Removing mean of each channel ...</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Final training data range: -456.566 to 562.934</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Computing the sphering
matrix...</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Starting weights are the identity matrix ...</font></div><div style="background-color:transparent">
<font face="arial, helvetica, sans-serif" style="font-style:italic">Sphering the data ...</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Beginning ICA training ...</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Lowering learning rate to 0.0009 and starting again.</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">step 1 - lrate 0.000900, wchange 141.51904569, angledelta 0.0 deg</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Lowering learning rate to 0.00081 and starting again.</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">step 1 - lrate 0.000810, wchange 159.49273116, angledelta 0.0 deg</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Lowering learning rate to 0.000729 and starting again.</font></div><div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">step 1 - lrate 0.000729, wchange <a href="tel:143.22748820" value="+14322748820" target="_blank">143.22748820</a>, angledelta 0.0 deg</font></div>
<div style="background-color:transparent"><font face="arial, helvetica, sans-serif" style="font-style:italic">Lowering learning rate to 0.0006561 and starting again.</font></div><div style="background-color:transparent">
<font face="arial, helvetica, sans-serif" style="font-style:italic">step 1 - lrate 0.000656, wchange 136.88732118, angledelta 0.0 deg</font></div>
</div></div></div><br>_______________________________________________<br>
Eeglablist page: <a href="http://sccn.ucsd.edu/eeglab/eeglabmail.html" target="_blank">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>
To unsubscribe, send an empty email to <a href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu" target="_blank">eeglablist-unsubscribe@sccn.ucsd.edu</a><br>
For digest mode, send an email with the subject "set digest mime" to <a href="mailto:eeglablist-request@sccn.ucsd.edu" target="_blank">eeglablist-request@sccn.ucsd.edu</a><br></blockquote></div><br><br clear="all">
<div><br></div>
-- <br><div dir="ltr">Makoto Miyakoshi<br>Swartz Center for Computational Neuroscience<br>Institute for Neural Computation, University of California San Diego<br></div>
</div></div>