cnn_cifar_init.m

function net = cnn_cifar_init(opts)

lr = [.1 2] ;
percentN = 4 ;

% Define network CIFAR10-quick
net.layers = {} ;

% Block 1
net.layers{end+1} = struct('type', 'conv', ...
                           'weights', {{0.01*randn(5,5,3,32, 'single'), zeros(1, 32, 'single')}}, ...
                           'learningRate', lr, ...
                           'stride', 1, ...
                           'pad', 2) ;
net.layers{end+1} = struct('type', 'pool', ...
                           'method', 'max', ...
                           'pool', [3 3], ...
                           'stride', 2, ...
                           'pad', [0 1 0 1]) ;
net.layers{end+1} = struct('type', 'mixlu', 'percentNeg',percentN) ;

% Block 2
net.layers{end+1} = struct('type', 'conv', ...
                           'weights', {{0.05*randn(5,5,32,32, 'single'), zeros(1,32,'single')}}, ...
                           'learningRate', lr, ...
                           'stride', 1, ...
                           'pad', 2) ;
net.layers{end+1} = struct('type', 'mixlu', 'percentNeg', percentN) ;
net.layers{end+1} = struct('type', 'pool', ...
                           'method', 'avg', ...
                           'pool', [3 3], ...
                           'stride', 2, ...
                           'pad', [0 1 0 1]) ; % Emulate caffe

% Block 3
net.layers{end+1} = struct('type', 'conv', ...
                           'weights', {{0.05*randn(5,5,32,64, 'single'), zeros(1,64,'single')}}, ...
                           'learningRate', lr, ...
                           'stride', 1, ...
                           'pad', 2) ;
net.layers{end+1} = struct('type', 'mixlu', 'percentNeg', percentN) ;
net.layers{end+1} = struct('type', 'pool', ...
                           'method', 'avg', ...
                           'pool', [3 3], ...
                           'stride', 2, ...
                           'pad', [0 1 0 1]) ; % Emulate caffe

% Block 4
net.layers{end+1} = struct('type', 'conv', ...
                           'weights', {{0.05*randn(4,4,64,64, 'single'), zeros(1,64,'single')}}, ...
                           'learningRate', lr, ...
                           'stride', 1, ...
                           'pad', 0) ;
net.layers{end+1} = struct('type', 'mixlu', 'percentNeg', percentN) ;

% Block 5
net.layers{end+1} = struct('type', 'conv', ...
                           'weights', {{0.05*randn(1,1,64,10, 'single'), zeros(1,10,'single')}}, ...
                           'learningRate', .1*lr, ...
                           'stride', 1, ...
                           'pad', 0) ;

% Loss layer
net.layers{end+1} = struct('type', 'softmaxloss') ;