Hello list,<br><br>I'm with a problem running the i.gensig. <br>I do:<br><br>i.gensig trainingmap=areasraster@Portugal_9 group=grupo@Portugal_9 subgroup=subgrupo signaturefile=assinaturascomareas<br><br>where areasrastes has 4 classes. <br>
The *r.stats -c input=areasraster@Portugal_9* is <br>1 92<br>2 234<br>3 2434<br>4 3016<br>* 287024 <br><br>And I get the following:<br><br>Finding training classes...<br>4 classes found<br>Calculating class means...<br>Calculating class covariance matrices...<br>
Signature 1 not invertible<br>Signature 2 not invertible<br>Signature 3 not invertible<br>Signature 4 not invertible<br>i.gensig complete.<br><br>My question is: do I need any package to deal with<br>matrix calculus? If so, each? If not, what is wrong <br>
here? <br><br>Thanks in advance<br>JD<br><br><br>