Comments on: Logistic regression for multi-class classification problems – a vectorized MATLAB/Octave approach
http://georgepavlides.info/logistic-regression-multi-class-classification-problems-vectorized-matlab-octave-approach/
More or less about me and everything that I learn, or hear, or think and find interesting to note.Fri, 09 Feb 2018 13:42:31 +0000hourly1https://wordpress.org/?v=4.9.4By: Matrix-based implementation of neural network back-propagation training – a MATLAB/Octave approach | @sepdek's
http://georgepavlides.info/logistic-regression-multi-class-classification-problems-vectorized-matlab-octave-approach/comment-page-1/#comment-31031
Fri, 09 Feb 2018 13:42:31 +0000http://georgepavlides.info/?p=1974#comment-31031[…] the gradient of the function that defines that cost. Similarly to linear or logistic regression (presented in a previous post), it can be seen as a pure cost function minimisation method, in which the definition of the cost […]
]]>