<div dir="ltr"><p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif;background-image:initial;background-repeat:initial">Hey guys, I'm having
problems wrapping my head around this.</span><span lang="EN-US" style="font-size:12pt;font-family:"Times New Roman",serif"></span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">I'm trying to run a representational similarity
analysis RSA, on the same data in which I run a MVPA analysis to compare both
outputs. My problem is that when I introduce the data for MVPA, I use feature
selection like in the tutorial, so I grab 5% of the ANOVA’s output and use it
to train and test the data. But in the tutorial, seems like you grab all the
voxels, not only a selection.</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">How can I grab only the same 5% for RSA?</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">This is the code I use for MVPA:</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">clf = LinearCSVMC()</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">fsel =
SensitivityBasedFeatureSelection(OneWayAnova(), FractionTailSelector(0.05,
mode='select', tail='upper'))</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">fclf = FeatureSelectionClassifier(clf, fsel)</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">cvte = CrossValidation(fclf, NFoldPartitioner(),
errorfx=lambda p, t: np.mean(p == t), enable_ca=['stats'])</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">cv_results = cvte(ds)</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">This is the one I got for RSA:</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif"> </span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">dsm = rsa.PDist(square=True)</span></p>

<p class="MsoNormal" style="margin-bottom:0.0001pt"><span lang="EN-US" style="font-size:12pt;font-family:Arial,sans-serif">res = dsm(ds)</span></p></div>