help
help > Possible bug in hmrMotionCorrectPCA
Jul 10, 2018 08:07 PM | Frank Fishburn
Possible bug in hmrMotionCorrectPCA
Hello,
I am testing tPCA on some simulated data and am finding that nSV sometimes gets set to 0 when the first singular value is greater than nSV, resulting in no correction being applied. I don't think this is a common scenario, so it's probably not a major problem. However, it looks to me like the issue is on lines 67-70:
if nSV<1 & nSV>0
ev = diag(svsc < nSV);
nSV = find(diag(ev)==0,1)-1;
end
It looks to me like it can be fixed with this change:
if nSV<1 & nSV>0
nSV = find(svsc>=nSV,1,'first');
end
Hope this helps. Apologies if I missed something.
Best,
Frank
I am testing tPCA on some simulated data and am finding that nSV sometimes gets set to 0 when the first singular value is greater than nSV, resulting in no correction being applied. I don't think this is a common scenario, so it's probably not a major problem. However, it looks to me like the issue is on lines 67-70:
if nSV<1 & nSV>0
ev = diag(svsc < nSV);
nSV = find(diag(ev)==0,1)-1;
end
It looks to me like it can be fixed with this change:
if nSV<1 & nSV>0
nSV = find(svsc>=nSV,1,'first');
end
Hope this helps. Apologies if I missed something.
Best,
Frank