I'm having a similar problem whereby every time I run i.gensig followed by i.maxlik it results in a single class for all the pixels. This doesn't happen when I use i.cluster with i.maxlik. I haven't analysed the sig file closely to see if the problem might be there, but I will try that tomorrow. At any rate, perhaps this is related to the problem Helena and Antonio are reporting.<div>
<br></div><div>- Nick<br><br><div class="gmail_quote">On Wed, Nov 17, 2010 at 6:51 PM, Helena Herrera <span dir="ltr"><<a href="mailto:helenaherrera1980@gmail.com">helenaherrera1980@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
I had exacly the same problem. My solution was, by-hand, creating a reclass file and then run this.<div><br></div><div>But, as far as I can see this is a major failure of i.gensig and Classification methdology. Or are we (me and Antonio) seeing this wrong? because, if training map sets values as 1000 or 2 or 3, they shold be kept in signature file, in order to be used in i.maxlik</div>
<div><br></div><div>has anyone any solution?</div><div>Helena</div><font color="#888888"><div><br></div>
</font><br>_______________________________________________<br>
grass-user mailing list<br>
<a href="mailto:grass-user@lists.osgeo.org">grass-user@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/grass-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/grass-user</a><br>
<br></blockquote></div><br></div>