Wed, 05 May 2010 10:24:54 +0100 solved the problem with equivariance by first eta-normalising the goal
Christian Urban <urbanc@in.tum.de> [Wed, 05 May 2010 10:24:54 +0100] rev 2064
solved the problem with equivariance by first eta-normalising the goal
(0) -1000 -300 -100 -30 -10 -1 +1 +10 +30 +100 +300 +1000 tip