<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi!<div><br><div><div>On 21 May 2008, at 09:20, ilya wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><p style="margin: 0.0px 0.0px 0.0px 0.0px"><font face="Helvetica" size="3" style="font: 12.0px Helvetica">As I understand conjugate gradient methods are not suitable for TS</font></p> <p style="margin: 0.0px 0.0px 0.0px 0.0px"><font face="Helvetica" size="3" style="font: 12.0px Helvetica">optimizations? Or I am wrong.</font></p></blockquote>Yes.. you're wrong..</div><div>instead of the "plug and pray" algorithm why don't you have a look at the regtest files?</div><div>this will make you save a considerable amount of time..</div><div><br><blockquote type="cite"> <p style="margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica; min-height: 14.0px"><br></p> <p style="margin: 0.0px 0.0px 0.0px 0.0px"><font face="Helvetica" size="3" style="font: 12.0px Helvetica">And the main question: Is TS location algorithm implemented (the input</font></p> <p style="margin: 0.0px 0.0px 0.0px 0.0px"><font face="Helvetica" size="3" style="font: 12.0px Helvetica">section is present so I guess yes??) ?</font></p> </blockquote>Yes! it is implemented and we are still fully validating it..</div><div>Of course CG are not the best strategy.. LBFGS would perform much better..</div><div>but at the moment that's it.</div><div><br class="webkit-block-placeholder"></div><div>Teo</div><br></div><div>p.s.: let me stress one important thing.. a section may even be present but there's no</div><div>guarantee that the method works or is bug free.. </div></body></html>