<div dir="ltr">Hi,<div><br></div><div class="gmail_extra"><div class="gmail_quote">On Mon, Sep 28, 2015 at 4:42 PM, Richard Dinga <span dir="ltr"><<a href="mailto:dinga92@gmail.com" target="_blank">dinga92@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><span class=""><pre>> I'm performing searchlight support vector regression, not classification,
> so my error goes from 0 to 2 instead of 0 to 1. Can I simply take `2 -
> error` (so higher is better) or does the algorithm require the values to
> lie between 0 and 1?<br></pre></span><pre>You can do that, as long as higher is better. Algorithm will just sort them and find nth biggest value.<span class=""><br><br>> The docs say the accuracy maps must be the result of
> classification, but is there a specific reason regression won't work?<br><br></span></pre><pre>I have no idea :) Original paper doesn't mention regression. Will p-values and FDR still make sense?<br></pre></div>
</blockquote></div>Right now I cannot think of a reason we it wouldn't work -- as long as "bigger is better".</div><div class="gmail_extra"><br></div><div class="gmail_extra">The algorithm converts pretty much anything into probabilities/frequencies. 1. Value corresponding to a particular probability under H0 -> threshold. 2. frequency with which a particular blob size has been observered in a group average map under H0.</div><div class="gmail_extra"><br></div><div class="gmail_extra">Should work.</div><div class="gmail_extra"><br></div><div class="gmail_extra">Michael</div><div class="gmail_extra"><br></div><div class="gmail_extra"><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div>Michael Hanke</div><div><a href="http://mih.voxindeserto.de" target="_blank">http://mih.voxindeserto.de</a></div><div><div><br></div></div></div></div>
</div></div>