CSE546 submission

nist.cfg

@@ -2,7 +2,7 @@
 width=28
 height=28
 channels=1
-filters=6
+filters=5
 size=5
 stride=1
 activation=ramp
@@ -11,7 +11,7 @@ activation=ramp
 stride=2
 
 [conv]
-filters=16
+filters=50
 size=5
 stride=1
 activation=ramp
@@ -19,17 +19,8 @@ activation=ramp
 [maxpool]
 stride=2
 
-[conv]
-filters=120
-size=3
-stride=1
-activation=ramp
-
-[maxpool]
-stride=2
-
 [conn]
-output = 80
+output = 100
 activation=ramp
 
 [conn]

src/convolutional_layer.c

@@ -39,7 +39,7 @@ convolutional_layer *make_convolutional_layer(int h, int w, int c, int n, int si
     layer->w = w;
     layer->c = c;
     layer->n = n;
-    layer->edge = 1;
+    layer->edge = 0;
     layer->stride = stride;
     layer->kernels = calloc(n, sizeof(image));
     layer->kernel_updates = calloc(n, sizeof(image));

src/network.c

@@ -63,7 +63,7 @@ void update_network(network net, double step, double momentum, double decay)
         }
         else if(net.types[i] == CONNECTED){
             connected_layer layer = *(connected_layer *)net.layers[i];
-            update_connected_layer(layer, step, momentum, decay);
+            update_connected_layer(layer, step, momentum, 0);
         }
     }
 }

src/tests.c

@@ -195,26 +195,31 @@ void test_full()
 void test_nist()
 {
     srand(444444);
+    srand(888888);
     network net = parse_network_cfg("nist.cfg");
     data train = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
     data test = load_categorical_data_csv("mnist/mnist_test.csv",0,10);
     normalize_data_rows(train);
     normalize_data_rows(test);
-    randomize_data(train);
+    //randomize_data(train);
     int count = 0;
     double lr = .0005;
-    while(++count <= 1){
+    double momentum = .9;
-        double acc = train_network_sgd(net, train, 10000, lr, .9, .001);
+    double decay = 0.01;
-        printf("Training Accuracy: %lf\n", acc);
+    while(++count <= 1000){
-        lr /= 2; 
+        double acc = train_network_sgd(net, train, 1000, lr, momentum, decay);
+        printf("Training Accuracy: %lf, Params: %f %f %f\n", acc, lr, momentum, decay);
+        visualize_network(net);
+        cvWaitKey(100);
+        //lr /= 2; 
+        if(count%5 == 0 && 0){
+            double train_acc = network_accuracy(net, train);
+            fprintf(stderr, "\nTRAIN: %f\n", train_acc);
+            double test_acc = network_accuracy(net, test);
+            fprintf(stderr, "TEST: %f\n\n", test_acc);
+            printf("%d, %f, %f\n", count, train_acc, test_acc);
+        }
     }
-    double train_acc = network_accuracy(net, train);
-    fprintf(stderr, "\nTRAIN: %f\n", train_acc);
-    double test_acc = network_accuracy(net, test);
-    fprintf(stderr, "TEST: %f\n\n", test_acc);
-    printf("%d, %f, %f\n", count, train_acc, test_acc);
-    //end = clock();
-    //printf("Neural Net Learning: %lf seconds\n", (double)(end-start)/CLOCKS_PER_SEC);
 }
 
 void test_ensemble()
@@ -223,24 +228,25 @@ void test_ensemble()
     srand(888888);
     data d = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
     normalize_data_rows(d);
-    randomize_data(d);
     data test = load_categorical_data_csv("mnist/mnist_test.csv", 0,10);
     normalize_data_rows(test);
     data train = d;
     /*
-    data *split = split_data(d, 1, 10);
+       data *split = split_data(d, 1, 10);
-    data train = split[0];
+       data train = split[0];
-    data test = split[1];
+       data test = split[1];
-    */
+     */
     matrix prediction = make_matrix(test.y.rows, test.y.cols);
     int n = 30;
     for(i = 0; i < n; ++i){
         int count = 0;
         double lr = .0005;
+        double momentum = .9;
+        double decay = .01;
         network net = parse_network_cfg("nist.cfg");
-        while(++count <= 5){
+        while(++count <= 15){
-            double acc = train_network_sgd(net, train, train.X.rows, lr, .9, .001);
+            double acc = train_network_sgd(net, train, train.X.rows, lr, momentum, decay);
-            printf("Training Accuracy: %lf\n", acc);
+            printf("Training Accuracy: %lf Learning Rate: %f Momentum: %f Decay: %f\n", acc, lr, momentum, decay );
             lr /= 2; 
         }
         matrix partial = network_predict_data(net, test);