Resizing Detector: random=1 (/1.4 - x1.4), random=3 (/3.0 - x3.0), random=0 (no resizing). Added param: resize_step=32 to the [net] section: use resize_step=128 for 7 subsampling layers.

include/darknet.h

@@ -305,7 +305,7 @@ struct layer {
     float mask_scale;
     float class_scale;
     int bias_match;
-    int random;
+    float random;
     float ignore_thresh;
     float truth_thresh;
     float iou_thresh;
@@ -664,6 +664,7 @@ typedef struct network {
     int blur;
     int mixup;
     float label_smooth_eps;
+    int resize_step;
     int letter_box;
     float angle;
     float aspect;

src/detector.c

@@ -168,19 +168,22 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
     //while(i*imgs < N*120){
     while (get_current_batch(net) < net.max_batches) {
         if (l.random && count++ % 10 == 0) {
-            printf("Resizing\n");
+            float rand_coef = 1.4;
-            float random_val = rand_scale(1.4);    // *x or /x
+            if (l.random != 1.0) rand_coef = l.random;
-            int dim_w = roundl(random_val*init_w / 32 + 1) * 32;
+            printf("Resizing, random_coef = %.2f \n", rand_coef);
-            int dim_h = roundl(random_val*init_h / 32 + 1) * 32;
+            float random_val = rand_scale(rand_coef);    // *x or /x
+            int dim_w = roundl(random_val*init_w / net.resize_step + 1) * net.resize_step;
+            int dim_h = roundl(random_val*init_h / net.resize_step + 1) * net.resize_step;
+            if (random_val < 1 && (dim_w > init_w || dim_h > init_h)) dim_w = init_w, dim_h = init_h;
 
             // at the beginning
             if (avg_loss < 0) {
-                dim_w = roundl(1.4*init_w / 32 + 1) * 32;
+                dim_w = roundl(rand_coef*init_w / net.resize_step + 1) * net.resize_step;
-                dim_h = roundl(1.4*init_h / 32 + 1) * 32;
+                dim_h = roundl(rand_coef*init_h / net.resize_step + 1) * net.resize_step;
             }
 
-            if (dim_w < 32) dim_w = 32;
+            if (dim_w < net.resize_step) dim_w = net.resize_step;
-            if (dim_h < 32) dim_h = 32;
+            if (dim_h < net.resize_step) dim_h = net.resize_step;
 
             printf("%d x %d \n", dim_w, dim_h);
             args.w = dim_w;

src/parser.c

@@ -439,7 +439,7 @@ layer parse_yolo(list *options, size_params params)
     l.ignore_thresh = option_find_float(options, "ignore_thresh", .5);
     l.truth_thresh = option_find_float(options, "truth_thresh", 1);
     l.iou_thresh = option_find_float_quiet(options, "iou_thresh", 1); // recommended to use iou_thresh=0.213 in [yolo]
-    l.random = option_find_int_quiet(options, "random", 0);
+    l.random = option_find_float_quiet(options, "random", 0);
 
     char *map_file = option_find_str(options, "map", 0);
     if (map_file) l.map = read_map(map_file);
@@ -541,7 +541,7 @@ layer parse_gaussian_yolo(list *options, size_params params) // Gaussian_YOLOv3
     l.ignore_thresh = option_find_float(options, "ignore_thresh", .5);
     l.truth_thresh = option_find_float(options, "truth_thresh", 1);
     l.iou_thresh = option_find_float_quiet(options, "iou_thresh", 1); // recommended to use iou_thresh=0.213 in [yolo]
-    l.random = option_find_int_quiet(options, "random", 0);
+    l.random = option_find_float_quiet(options, "random", 0);
 
     char *map_file = option_find_str(options, "map", 0);
     if (map_file) l.map = read_map(map_file);
@@ -590,7 +590,7 @@ layer parse_region(list *options, size_params params)
     l.thresh = option_find_float(options, "thresh", .5);
     l.classfix = option_find_int_quiet(options, "classfix", 0);
     l.absolute = option_find_int_quiet(options, "absolute", 0);
-    l.random = option_find_int_quiet(options, "random", 0);
+    l.random = option_find_float_quiet(options, "random", 0);
 
     l.coord_scale = option_find_float(options, "coord_scale", 1);
     l.object_scale = option_find_float(options, "object_scale", 1);
@@ -639,7 +639,7 @@ detection_layer parse_detection(list *options, size_params params)
     layer.noobject_scale = option_find_float(options, "noobject_scale", 1);
     layer.class_scale = option_find_float(options, "class_scale", 1);
     layer.jitter = option_find_float(options, "jitter", .2);
-    layer.random = option_find_int_quiet(options, "random", 0);
+    layer.random = option_find_float_quiet(options, "random", 0);
     layer.reorg = option_find_int_quiet(options, "reorg", 0);
     return layer;
 }
@@ -1060,6 +1060,7 @@ void parse_net_options(list *options, network *net)
     else if (mosaic) net->mixup = 3;
     net->letter_box = option_find_int_quiet(options, "letter_box", 0);
     net->label_smooth_eps = option_find_float_quiet(options, "label_smooth_eps", 0.0f);
+    net->resize_step = option_find_float_quiet(options, "resize_step", 32);
 
     net->angle = option_find_float_quiet(options, "angle", 0);
     net->aspect = option_find_float_quiet(options, "aspect", 1);