[ADD] Documentation

Author: Jeffresh

Diff for: train.m

@@ -2,21 +2,17 @@
 function [neural_net,y_predicted, learning_rate]= train(neural_net,X,Y,...
 cost_function,learning_rate)
 
-%TRAIN Summary of this function goes here
-%   Detailed explanation goes here
+%TRAIN Training based in two steps, forwarding and backpropagation.
 
 out(1,1:2)={nan,X};
 
-% forward, get z(output) values of all layers.
-
-out = fordwarding(neural_net, out);
-
-%back propagation, modify the weights of the nn.
+% calculate de "error" of the last layer.
+out = forwarding(neural_net, out);
 
+% "propagate" that error to modify the weights of every layer
 neural_net = backpropagation(neural_net,cost_function, out, learning_rate, Y);  
 
-% la salida se almacena en la última posición de out.
-
+% usefull to calculate and analize the behavior of the cost in every epoch. (to detect overfitting, underfitting...)
 y_predicted = out{end,2};
 
 end