Solved conflict.

Author: albertoandreottiATgmail

tests/test_example_CNN.m

@@ -9,17 +9,13 @@
 %% ex1 Train a 6c-2s-12c-2s Convolutional neural network 
 %will run 1 epoch in about 200 second and get around 11% error. 
 %With 100 epochs you'll get around 1.2% error
-<<<<<<< HEAD
 
 if !isOctave() 
    rng(0)
 else
    rand('state',0)
 end
 
-=======
-rand('state',0)
->>>>>>> rasmus/master
 cnn.layers = {
     struct('type', 'i') %input layer
     struct('type', 'c', 'outputmaps', 6, 'kernelsize', 5) %convolution layer