For backpropagation algorithm, is it true to have weight, $w$ transposed in the expression of $dz^{[1]} = w^{[2]T}dz^{[2]} * g^{[1]'}(z^{[1]}) $ ? Could anyone show me why $w^{[2]T}$ instead of just $w$ ?
I have read other articles, but the transposed dimension of $w$ is ignored when chain deriv...
Umm yeh I like it. Although I'm still trying to understand how to structure my stuff.. I've got some code which I think should be in a module but not sure
Basically, before I run anything, I need to check that the current branch has a dedicated static directory on my server where I plan on saving any data that needs to persist which is too large or not suitable for Git