why we want large (log y cap) when y=1… the term is (-log(ycap)) so our objective is to make loss function to be as minimum possible ( try to be make loss function = 0 ) so that is possible only when when we do (log (y cap )) as small as possible so that we get less negative term ?? - (reference - 5:31 (video name- logistic regression cost function) )
Welcome to the community !
We have several discussions in here, which can be searchable from a magnifying glass icon on top-
right. You may be able to find a similar issue quickly without waiting a response from someone.
There is a thread answered by Paul and Elemento. Please visit here.