neingeist/coursera-ml-007-exercises
Archived
1
0
Fork 0
Code Releases Activity

Random initialization

Browse source
This commit is contained in:
neingeist 2014-11-01 21:22:34 +01:00
parent 863f1d7157
commit 052f0625c3
1 changed files with 6 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

17
ex4/randInitializeWeights.m
View file

@ -1,15 +1,15 @@
function W = randInitializeWeights(L_in, L_out) function W = randInitializeWeights(L_in, L_out)
%RANDINITIALIZEWEIGHTS Randomly initialize the weights of a layer with L_in %RANDINITIALIZEWEIGHTS Randomly initialize the weights of a layer with L_in
%incoming connections and L_out outgoing connections %incoming connections and L_out outgoing connections
% W = RANDINITIALIZEWEIGHTS(L_in, L_out) randomly initializes the weights % W = RANDINITIALIZEWEIGHTS(L_in, L_out) randomly initializes the weights
% of a layer with L_in incoming connections and L_out outgoing % of a layer with L_in incoming connections and L_out outgoing
% connections. % connections.
% %
% Note that W should be set to a matrix of size(L_out, 1 + L_in) as % Note that W should be set to a matrix of size(L_out, 1 + L_in) as
% the column row of W handles the "bias" terms % the column row of W handles the "bias" terms
% %
% You need to return the following variables correctly % You need to return the following variables correctly
W = zeros(L_out, 1 + L_in); W = zeros(L_out, 1 + L_in);
% ====================== YOUR CODE HERE ====================== % ====================== YOUR CODE HERE ======================
@ -19,13 +19,8 @@ W = zeros(L_out, 1 + L_in);
% Note: The first row of W corresponds to the parameters for the bias units % Note: The first row of W corresponds to the parameters for the bias units
% %
epsilon_init = 0.12;
W = rand(L_out, 1 + L_in) * 2 * epsilon_init - epsilon_init;
% ========================================================================= % =========================================================================