--- a
+++ b/combinedDeepLearningActiveContour/minFunc/logistic/LogisticHv.m
@@ -0,0 +1,8 @@
+function [Hv] = LogisticHv(v,w,X,y)
+% v(feature,1) - vector that we will multiply Hessian by
+% w(feature,1)
+% X(instance,feature)
+% y(instance,1)
+
+sig = 1./(1+exp(-y.*(X*w)));
+Hv = X.'*(sig.*(1-sig).*(X*v));