logregDist.demoLaplaceGirolami laplace approximation to posterior

blutooth / pmtk

Automatically exported from code.google.com/p/pmtk

0 stars 0 forks source link

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago

Looks like the distribution is centered around the wrong point.

Original issue reported on code.google.com by mattbdun...@gmail.com on 27 Oct 2008 at 3:23

GoogleCodeExporter commented 8 years ago

Bug introduced when logregDist was combined with multinomLogregDist

Original comment by mattbdun...@gmail.com on 27 Oct 2008 at 3:29

GoogleCodeExporter commented 8 years ago

Somehow -wMAP is used instead of wMAP, by using -wMAP in plotting laplace 
approx to 
the posterior, the output looks correct.

Original comment by mattbdun...@gmail.com on 27 Oct 2008 at 6:37