octave compatibility fixes

Author: rasmusbergpalm

NN/nnchecknumgrad.m

@@ -8,9 +8,9 @@ function nnchecknumgrad(nn, x, y)
                 nn_m = nn; nn_p = nn;
                 nn_m.W{l}(i, j) = nn.W{l}(i, j) - epsilon;
                 nn_p.W{l}(i, j) = nn.W{l}(i, j) + epsilon;
-                rng(0);
+				rand('state',0)
                 nn_m = nnff(nn_m, x, y);
-                rng(0);
+                rand('state',0)
                 nn_p = nnff(nn_p, x, y);
                 dW = (nn_p.L - nn_m.L) / (2 * epsilon);
                 e = abs(dW - nn.dW{l}(i, j));

NN/nnupdatefigures.m

@@ -43,40 +43,21 @@ function nnupdatefigures(nn,fhandle,L,opts,i)
         legend(p1, M,'Location','NorthEast');
         set(p1, 'Xlim',[0,opts.numepochs + 1])
 
-        if i ==2 % speeds up plotting by factor of ~2
-            set(gca,'LegendColorbarListeners',[]);
-            setappdata(gca,'LegendColorbarManualSpace',1);
-            setappdata(gca,'LegendColorbarReclaimSpace',1);
-        end
-               
         p2 = subplot(1,2,2);
         plot(plot_x,plot_yfrac);
         xlabel('Number of epochs'); ylabel('Misclassification rate');
         title('Misclassification rate')
         legend(p2, M,'Location','NorthEast');
         set(p2, 'Xlim',[0,opts.numepochs + 1])
 
-        if i ==2 % speeds up plotting by factor of ~2
-            set(gca,'LegendColorbarListeners',[]);
-            setappdata(gca,'LegendColorbarManualSpace',1);
-            setappdata(gca,'LegendColorbarReclaimSpace',1);
-        end
-
     else
 
         p = plot(plot_x,plot_ye);
         xlabel('Number of epochs'); ylabel('Error');title('Error');
         legend(p, M,'Location','NorthEast');
         set(gca, 'Xlim',[0,opts.numepochs + 1])
-        
-        if i ==2 % speeds up plotting by factor of ~2
-            set(gca,'LegendColorbarListeners',[]);
-            setappdata(gca,'LegendColorbarManualSpace',1);
-            setappdata(gca,'LegendColorbarReclaimSpace',1);
-            
-        end
-        
+                
     end
     drawnow;
 end
-end
+end

tests/test_example_CNN.m

@@ -9,7 +9,7 @@
 %% 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
-rng(0)
+rand('state',0)
 cnn.layers = {
     struct('type', 'i') %input layer
     struct('type', 'c', 'outputmaps', 6, 'kernelsize', 5) %convolution layer

tests/test_example_DBN.m

@@ -7,7 +7,7 @@
 test_y  = double(test_y);
 
 %%  ex1 train a 100 hidden unit RBM and visualize its weights
-rng(0);
+rand('state',0)
 dbn.sizes = [100];
 opts.numepochs =   1;
 opts.batchsize = 100;
@@ -18,7 +18,7 @@
 figure; visualize(dbn.rbm{1}.W');   %  Visualize the RBM weights
 
 %%  ex2 train a 100-100 hidden unit DBN and use its weights to initialize a NN
-rng(0);
+rand('state',0)
 %train dbn
 dbn.sizes = [100 100];
 opts.numepochs =   1;

tests/test_example_NN.m

@@ -11,7 +11,7 @@
 test_x = normalize(test_x, mu, sigma);
 
 %% ex1 vanilla neural net
-rng(0);
+rand('state',0)
 nn = nnsetup([784 100 10]);
 opts.numepochs =  1;   %  Number of full sweeps through data
 opts.batchsize = 100;  %  Take a mean gradient step over this many samples
@@ -21,16 +21,8 @@
 
 assert(er < 0.08, 'Too big error');
 
-% Make an artificial one and verify that we can predict it
-x = zeros(1,28,28);
-x(:, 14:15, 6:22) = 1;
-x = reshape(x,1,28^2);
-figure; visualize(x');
-predicted = nnpredict(nn,x)-1;
-
-assert(predicted == 1);
 %% ex2 neural net with L2 weight decay
-rng(0);
+rand('state',0)
 nn = nnsetup([784 100 10]);
 
 nn.weightPenaltyL2 = 1e-4;  %  L2 weight decay
@@ -44,7 +36,7 @@
 
 
 %% ex3 neural net with dropout
-rng(0);
+rand('state',0)
 nn = nnsetup([784 100 10]);
 
 nn.dropoutFraction = 0.5;   %  Dropout fraction 
@@ -57,7 +49,7 @@
 assert(er < 0.1, 'Too big error');
 
 %% ex4 neural net with sigmoid activation function
-rng(0);
+rand('state',0)
 nn = nnsetup([784 100 10]);
 
 nn.activation_function = 'sigm';    %  Sigmoid activation function
@@ -71,7 +63,7 @@
 assert(er < 0.1, 'Too big error');
 
 %% ex5 plotting functionality
-rng(0);
+rand('state',0)
 nn = nnsetup([784 20 10]);
 opts.numepochs         = 5;            %  Number of full sweeps through data
 nn.output              = 'softmax';    %  use softmax output
@@ -90,7 +82,7 @@
 vy   = train_y(1:10000,:);
 ty = train_y(10001:end,:);
 
-rng(0);
+rand('state',0)
 nn                      = nnsetup([784 20 10]);     
 nn.output               = 'softmax';                   %  use softmax output
 opts.numepochs          = 5;                           %  Number of full sweeps through data
@@ -99,4 +91,4 @@
 nn = nntrain(nn, tx, ty, opts, vx, vy);                %  nntrain takes validation set as last two arguments (optionally)
 
 [er, bad] = nntest(nn, test_x, test_y);
-assert(er < 0.1, 'Too big error');
+assert(er < 0.1, 'Too big error');

tests/test_example_SAE.m

@@ -8,7 +8,7 @@
 
 %%  ex1 train a 100 hidden unit SDAE and use it to initialize a FFNN
 %  Setup and train a stacked denoising autoencoder (SDAE)
-rng(0);
+rand('state',0)
 sae = saesetup([784 100]);
 sae.ae{1}.activation_function       = 'sigm';
 sae.ae{1}.learningRate              = 1;
@@ -33,7 +33,7 @@
 
 %% ex2 train a 100-100 hidden unit SDAE and use it to initialize a FFNN
 %  Setup and train a stacked denoising autoencoder (SDAE)
-rng(0);
+rand('state',0)
 sae = saesetup([784 100 100]);
 sae.ae{1}.activation_function       = 'sigm';
 sae.ae{1}.learningRate              = 1;

tests/test_nn_gradients_are_numerically_correct.m

@@ -17,10 +17,10 @@
             nn.output = output{1};
             nn.dropoutFraction = dropoutFraction{1};
 
-            rng(0)
+            rand('state',0)
             nn = nnff(nn, batch_x, y);
             nn = nnbp(nn);
             nnchecknumgrad(nn, batch_x, y);
         end
     end
-end
+end