function net = newLeNet5(inp_rows, inp_cols, n_classes, eta, momentum, parS2C3_cmat, parC5_nFMs)

if ~exist('inp_rows', 'var') || isempty(inp_rows), inp_rows = 32; end

if ~exist('inp_cols', 'var') || isempty(inp_cols), inp_cols = inp_rows; end

if ~exist('n_classes', 'var') || isempty(n_classes), n_classes = 10; end

if ~exist('eta', 'var') || isempty(eta), eta = 5e-4; end

if ~exist('momentum', 'var') || isempty(momentum), momentum = 0; end

if ~exist('parS2C3_cmat', 'var') || isempty(parS2C3_cmat),

S2C3_cmat{1} = [1 0 0 0 1 1 1 0 0 1 1 1 1 0 1 1

1 1 0 0 0 1 1 1 0 0 1 1 1 1 0 1

1 1 1 0 0 0 1 1 1 0 0 1 0 1 1 1

0 1 1 1 0 0 1 1 1 1 0 0 1 0 1 1

0 0 1 1 1 0 0 1 1 1 1 0 1 1 0 1

0 0 0 1 1 1 0 0 1 1 1 1 0 1 1 1]';

else

S2C3_cmat{1} = parS2C3_cmat;

end

if ~exist('parC5_nFMs', 'var') || isempty(parC5_nFMs), parC5_nFMs = 120; end

trainable = true;

none = [];

% C1 layer

C1_nFMs = size(S2C3_cmat{1}, 2);

C1_is = [inp_rows inp_cols];

C1_ks = [5 5];

C1_ss = [1 1];

C1_os = ceil((C1_is - C1_ks + 1) ./ C1_ss);

% S2 layer

S2_ks = [2 2];

S2_is = C1_os;

S2_ss = S2_ks;

S2_nFMs = C1_nFMs;

S2_os = ceil((S2_is - S2_ks + 1) ./ S2_ss);

% C3 layer

C3_nFMs = size(S2C3_cmat{1}, 1);

C3_is = S2_os;

C3_ks = [5 5];

C3_ss = [1 1];

C3_os = ceil((C3_is - C3_ks + 1) ./ C3_ss);

% S4 layer

S4_ks = [2 2];

S4_nFMs = C3_nFMs;

S4_is = C3_os;

S4_ss = S4_ks;

S4_os = ceil((S4_is - S4_ks + 1) ./ S4_ss);

% C5 layer

C5_ks = S4_os;

C5_nFMs = parC5_nFMs;

C5_is = S4_os;

C5_ss = [1 1];

C5_os = ceil((C5_is - C5_ks + 1) ./ C5_ss);

% F6 layer

F6_nFMs = n_classes;

% F6_is = C5_os;

F6_ks = C5_os;

F6_ss = [1 1];

% F6_os = ceil((F6_is - F6_ks + 1) ./ F6_ss); % == [1x1]

S4C5_cmat{1} = ones(C5_nFMs, S4_nFMs); % S4 is fully connected to C5

net = myCNN('LeNet', [inp_rows inp_cols], eta, momentum); % creates object

net = add_layer(net, 'C1', trainable, 'C', C1_nFMs, none, none, C1_ks, C1_ss);

net = add_layer(net, 'S2', trainable, 'S', S2_nFMs, none, none, S2_ks, S2_ss);

net = add_layer(net, 'C3', trainable, 'C', C3_nFMs, S2C3_cmat, none, C3_ks, C3_ss);

net = add_layer(net, 'S4', trainable, 'S', S4_nFMs, none, none, S4_ks, S4_ss);

net = add_layer(net, 'C5', trainable, 'C', C5_nFMs, S4C5_cmat, none, C5_ks, C5_ss);

net = add_layer(net, 'out', trainable, 'F', F6_nFMs, none, none, F6_ks, F6_ss);

net = update_stat(net);

net = init_net(net, 1, 2);

end % of function newLeNet5