So I have this new programming paradigm.......

Makefile

@@ -41,10 +41,10 @@ CFLAGS+= -DCUDNN
 LDFLAGS+= -lcudnn
 endif
 
-OBJ=gemm.o utils.o cuda.o deconvolutional_layer.o convolutional_layer.o list.o image.o activations.o im2col.o col2im.o blas.o crop_layer.o dropout_layer.o maxpool_layer.o softmax_layer.o data.o matrix.o network.o connected_layer.o cost_layer.o parser.o option_list.o darknet.o detection_layer.o captcha.o route_layer.o writing.o box.o nightmare.o normalization_layer.o avgpool_layer.o coco.o dice.o yolo.o detector.o layer.o compare.o classifier.o local_layer.o swag.o shortcut_layer.o activation_layer.o rnn_layer.o gru_layer.o rnn.o rnn_vid.o crnn_layer.o demo.o tag.o cifar.o go.o batchnorm_layer.o art.o region_layer.o reorg_layer.o super.o voxel.o
+OBJ=gemm.o utils.o cuda.o convolutional_layer.o list.o image.o activations.o im2col.o col2im.o blas.o crop_layer.o dropout_layer.o maxpool_layer.o softmax_layer.o data.o matrix.o network.o connected_layer.o cost_layer.o parser.o option_list.o darknet.o detection_layer.o captcha.o route_layer.o writing.o box.o nightmare.o normalization_layer.o avgpool_layer.o coco.o dice.o yolo.o detector.o layer.o compare.o classifier.o local_layer.o swag.o shortcut_layer.o activation_layer.o rnn_layer.o gru_layer.o rnn.o rnn_vid.o crnn_layer.o demo.o tag.o cifar.o go.o batchnorm_layer.o art.o region_layer.o reorg_layer.o super.o voxel.o
 ifeq ($(GPU), 1) 
 LDFLAGS+= -lstdc++ 
-OBJ+=convolutional_kernels.o deconvolutional_kernels.o activation_kernels.o im2col_kernels.o col2im_kernels.o blas_kernels.o crop_layer_kernels.o dropout_layer_kernels.o maxpool_layer_kernels.o softmax_layer_kernels.o network_kernels.o avgpool_layer_kernels.o
+OBJ+=convolutional_kernels.o activation_kernels.o im2col_kernels.o col2im_kernels.o blas_kernels.o crop_layer_kernels.o dropout_layer_kernels.o maxpool_layer_kernels.o softmax_layer_kernels.o network_kernels.o avgpool_layer_kernels.o
 endif
 
 OBJS = $(addprefix $(OBJDIR), $(OBJ))

data/labels/make_labels.py

@@ -1,6 +1,19 @@
 import os
+import string
+import pipes
 
-l = ["person","bicycle","car","motorcycle","airplane","bus","train","truck","boat","traffic light","fire hydrant","stop sign","parking meter","bench","bird","cat","dog","horse","sheep","cow","elephant","bear","zebra","giraffe","backpack","umbrella","handbag","tie","suitcase","frisbee","skis","snowboard","sports ball","kite","baseball bat","baseball glove","skateboard","surfboard","tennis racket","bottle","wine glass","cup","fork","knife","spoon","bowl","banana","apple","sandwich","orange","broccoli","carrot","hot dog","pizza","donut","cake","chair","couch","potted plant","bed","dining table","toilet","tv","laptop","mouse","remote","keyboard","cell phone","microwave","oven","toaster","sink","refrigerator","book","clock","vase","scissors","teddy bear","hair drier","toothbrush", "aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]
+#l = ["person","bicycle","car","motorcycle","airplane","bus","train","truck","boat","traffic light","fire hydrant","stop sign","parking meter","bench","bird","cat","dog","horse","sheep","cow","elephant","bear","zebra","giraffe","backpack","umbrella","handbag","tie","suitcase","frisbee","skis","snowboard","sports ball","kite","baseball bat","baseball glove","skateboard","surfboard","tennis racket","bottle","wine glass","cup","fork","knife","spoon","bowl","banana","apple","sandwich","orange","broccoli","carrot","hot dog","pizza","donut","cake","chair","couch","potted plant","bed","dining table","toilet","tv","laptop","mouse","remote","keyboard","cell phone","microwave","oven","toaster","sink","refrigerator","book","clock","vase","scissors","teddy bear","hair drier","toothbrush", "aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]
+
+l = string.printable
 
 for word in l:
-    os.system("convert -fill black -background white -bordercolor white -border 4 -font futura-normal -pointsize 18 label:\"%s\" \"%s.png\""%(word, word))
+    #os.system("convert -fill black -background white -bordercolor white -border 4 -font futura-normal -pointsize 18 label:\"%s\" \"%s.png\""%(word, word))
+    if word == ' ':
+        os.system('convert -fill black -background white -bordercolor white -font futura-normal -pointsize 64 label:"\ " 32.png')
+    elif word == '\\':
+        os.system('convert -fill black -background white -bordercolor white -font futura-normal -pointsize 64 label:"\\\\\\\\" 92.png')
+    elif ord(word) in [9,10,11,12,13,14]:
+        pass
+    else:
+        os.system("convert -fill black -background white -bordercolor white -font futura-normal -pointsize 64 label:%s \"%d.png\""%(pipes.quote(word), ord(word)))
+

src/activation_layer.c

@@ -21,7 +21,12 @@ layer make_activation_layer(int batch, int inputs, ACTIVATION activation)
     l.output = calloc(batch*inputs, sizeof(float*));
     l.delta = calloc(batch*inputs, sizeof(float*));
 
+    l.forward = forward_activation_layer;
+    l.backward = backward_activation_layer;
 #ifdef GPU
+    l.forward_gpu = forward_activation_layer_gpu;
+    l.backward_gpu = backward_activation_layer_gpu;
+
     l.output_gpu = cuda_make_array(l.output, inputs*batch);
     l.delta_gpu = cuda_make_array(l.delta, inputs*batch);
 #endif

src/art.c

@@ -8,6 +8,7 @@
 
 #ifdef OPENCV
 #include "opencv2/highgui/highgui_c.h"
+image get_image_from_stream(CvCapture *cap);
 #endif
 
 

src/avgpool_layer.c

@@ -19,7 +19,11 @@ avgpool_layer make_avgpool_layer(int batch, int w, int h, int c)
     int output_size = l.outputs * batch;
     l.output =  calloc(output_size, sizeof(float));
     l.delta =   calloc(output_size, sizeof(float));
+    l.forward = forward_avgpool_layer;
+    l.backward = backward_avgpool_layer;
     #ifdef GPU
+    l.forward_gpu = forward_avgpool_layer_gpu;
+    l.backward_gpu = backward_avgpool_layer_gpu;
     l.output_gpu  = cuda_make_array(l.output, output_size);
     l.delta_gpu   = cuda_make_array(l.delta, output_size);
     #endif

src/batchnorm_layer.c

@@ -28,7 +28,13 @@ layer make_batchnorm_layer(int batch, int w, int h, int c)
 
     layer.rolling_mean = calloc(c, sizeof(float));
     layer.rolling_variance = calloc(c, sizeof(float));
+
+    layer.forward = forward_batchnorm_layer;
+    layer.backward = backward_batchnorm_layer;
 #ifdef GPU
+    layer.forward_gpu = forward_batchnorm_layer_gpu;
+    layer.backward_gpu = backward_batchnorm_layer_gpu;
+
     layer.output_gpu =  cuda_make_array(layer.output, h * w * c * batch);
     layer.delta_gpu =   cuda_make_array(layer.delta, h * w * c * batch);
 

src/classifier.c

@@ -10,6 +10,7 @@
 
 #ifdef OPENCV
 #include "opencv2/highgui/highgui_c.h"
+image get_image_from_stream(CvCapture *cap);
 #endif
 
 list *read_data_cfg(char *filename)
@@ -57,25 +58,26 @@ void train_classifier_multi(char *datacfg, char *cfgfile, char *weightfile, int
 #ifdef GPU
     int i;
 
-    srand(time(0));
     float avg_loss = -1;
     char *base = basecfg(cfgfile);
     printf("%s\n", base);
     printf("%d\n", ngpus);
     network *nets = calloc(ngpus, sizeof(network));
+
+    srand(time(0));
+    int seed = rand();
     for(i = 0; i < ngpus; ++i){
+        srand(seed);
         cuda_set_device(gpus[i]);
         nets[i] = parse_network_cfg(cfgfile);
-        if(clear) *nets[i].seen = 0;
         if(weightfile){
             load_weights(&nets[i], weightfile);
         }
-    }
-    network net = nets[0];
-    for(i = 0; i < ngpus; ++i){
-        *nets[i].seen = *net.seen;
+        if(clear) *nets[i].seen = 0;
         nets[i].learning_rate *= ngpus;
     }
+    srand(time(0));
+    network net = nets[0];
 
     int imgs = net.batch * net.subdivisions * ngpus;
 

src/coco.c

@@ -12,14 +12,10 @@
 #include "opencv2/highgui/highgui_c.h"
 #endif
 
-void convert_detections(float *predictions, int classes, int num, int square, int side, int w, int h, float thresh, float **probs, box *boxes, int only_objectness);
-
 char *coco_classes[] = {"person","bicycle","car","motorcycle","airplane","bus","train","truck","boat","traffic light","fire hydrant","stop sign","parking meter","bench","bird","cat","dog","horse","sheep","cow","elephant","bear","zebra","giraffe","backpack","umbrella","handbag","tie","suitcase","frisbee","skis","snowboard","sports ball","kite","baseball bat","baseball glove","skateboard","surfboard","tennis racket","bottle","wine glass","cup","fork","knife","spoon","bowl","banana","apple","sandwich","orange","broccoli","carrot","hot dog","pizza","donut","cake","chair","couch","potted plant","bed","dining table","toilet","tv","laptop","mouse","remote","keyboard","cell phone","microwave","oven","toaster","sink","refrigerator","book","clock","vase","scissors","teddy bear","hair drier","toothbrush"};
 
 int coco_ids[] = {1,2,3,4,5,6,7,8,9,10,11,13,14,15,16,17,18,19,20,21,22,23,24,25,27,28,31,32,33,34,35,36,37,38,39,40,41,42,43,44,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,67,70,72,73,74,75,76,77,78,79,80,81,82,84,85,86,87,88,89,90};
 
-image coco_labels[80];
-
 void train_coco(char *cfgfile, char *weightfile)
 {
     //char *train_images = "/home/pjreddie/data/voc/test/train.txt";
@@ -160,7 +156,6 @@ void validate_coco(char *cfgfile, char *weightfile)
 
     layer l = net.layers[net.n-1];
     int classes = l.classes;
-    int square = l.sqrt;
     int side = l.side;
 
     int j;
@@ -217,10 +212,10 @@ void validate_coco(char *cfgfile, char *weightfile)
             char *path = paths[i+t-nthreads];
             int image_id = get_coco_image_id(path);
             float *X = val_resized[t].data;
-            float *predictions = network_predict(net, X);
+            network_predict(net, X);
             int w = val[t].w;
             int h = val[t].h;
-            convert_detections(predictions, classes, l.n, square, side, w, h, thresh, probs, boxes, 0);
+            get_detection_boxes(l, w, h, thresh, probs, boxes, 0);
             if (nms) do_nms_sort(boxes, probs, side*side*l.n, classes, iou_thresh);
             print_cocos(fp, image_id, boxes, probs, side*side*l.n, classes, w, h);
             free_image(val[t]);
@@ -250,7 +245,6 @@ void validate_coco_recall(char *cfgfile, char *weightfile)
 
     layer l = net.layers[net.n-1];
     int classes = l.classes;
-    int square = l.sqrt;
     int side = l.side;
 
     int j, k;
@@ -282,14 +276,15 @@ void validate_coco_recall(char *cfgfile, char *weightfile)
         image orig = load_image_color(path, 0, 0);
         image sized = resize_image(orig, net.w, net.h);
         char *id = basecfg(path);
-        float *predictions = network_predict(net, sized.data);
-        convert_detections(predictions, classes, l.n, square, side, 1, 1, thresh, probs, boxes, 1);
+        network_predict(net, sized.data);
+        get_detection_boxes(l, 1, 1, thresh, probs, boxes, 1);
         if (nms) do_nms(boxes, probs, side*side*l.n, 1, nms_thresh);
 
-        char *labelpath = find_replace(path, "images", "labels");
-        labelpath = find_replace(labelpath, "JPEGImages", "labels");
-        labelpath = find_replace(labelpath, ".jpg", ".txt");
-        labelpath = find_replace(labelpath, ".JPEG", ".txt");
+        char labelpath[4096];
+        find_replace(path, "images", "labels", labelpath);
+        find_replace(labelpath, "JPEGImages", "labels", labelpath);
+        find_replace(labelpath, ".jpg", ".txt", labelpath);
+        find_replace(labelpath, ".JPEG", ".txt", labelpath);
 
         int num_labels = 0;
         box_label *truth = read_boxes(labelpath, &num_labels);
@@ -323,7 +318,7 @@ void validate_coco_recall(char *cfgfile, char *weightfile)
 
 void test_coco(char *cfgfile, char *weightfile, char *filename, float thresh)
 {
-
+    image *alphabet = load_alphabet();
     network net = parse_network_cfg(cfgfile);
     if(weightfile){
         load_weights(&net, weightfile);
@@ -353,11 +348,11 @@ void test_coco(char *cfgfile, char *weightfile, char *filename, float thresh)
         image sized = resize_image(im, net.w, net.h);
         float *X = sized.data;
         time=clock();
-        float *predictions = network_predict(net, X);
+        network_predict(net, X);
         printf("%s: Predicted in %f seconds.\n", input, sec(clock()-time));
-        convert_detections(predictions, l.classes, l.n, l.sqrt, l.side, 1, 1, thresh, probs, boxes, 0);
+        get_detection_boxes(l, 1, 1, thresh, probs, boxes, 0);
         if (nms) do_nms_sort(boxes, probs, l.side*l.side*l.n, l.classes, nms);
-        draw_detections(im, l.side*l.side*l.n, thresh, boxes, probs, coco_classes, coco_labels, 80);
+        draw_detections(im, l.side*l.side*l.n, thresh, boxes, probs, coco_classes, alphabet, 80);
         save_image(im, "prediction");
         show_image(im, "predictions");
         free_image(im);
@@ -372,12 +367,7 @@ void test_coco(char *cfgfile, char *weightfile, char *filename, float thresh)
 
 void run_coco(int argc, char **argv)
 {
-    int i;
-    for(i = 0; i < 80; ++i){
-        char buff[256];
-        sprintf(buff, "data/labels/%s.png", coco_classes[i]);
-        coco_labels[i] = load_image_color(buff, 0, 0);
-    }
+    char *prefix = find_char_arg(argc, argv, "-prefix", 0);
     float thresh = find_float_arg(argc, argv, "-thresh", .2);
     int cam_index = find_int_arg(argc, argv, "-c", 0);
     int frame_skip = find_int_arg(argc, argv, "-s", 0);
@@ -394,5 +384,5 @@ void run_coco(int argc, char **argv)
     else if(0==strcmp(argv[2], "train")) train_coco(cfg, weights);
     else if(0==strcmp(argv[2], "valid")) validate_coco(cfg, weights);
     else if(0==strcmp(argv[2], "recall")) validate_coco_recall(cfg, weights);
-    else if(0==strcmp(argv[2], "demo")) demo(cfg, weights, thresh, cam_index, filename, coco_classes, coco_labels, 80, frame_skip);
+    else if(0==strcmp(argv[2], "demo")) demo(cfg, weights, thresh, cam_index, filename, coco_classes, 80, frame_skip, prefix);
 }

src/connected_layer.c

@@ -36,6 +36,10 @@ connected_layer make_connected_layer(int batch, int inputs, int outputs, ACTIVAT
     l.weights = calloc(outputs*inputs, sizeof(float));
     l.biases = calloc(outputs, sizeof(float));
 
+    l.forward = forward_connected_layer;
+    l.backward = backward_connected_layer;
+    l.update = update_connected_layer;
+
     //float scale = 1./sqrt(inputs);
     float scale = sqrt(2./inputs);
     for(i = 0; i < outputs*inputs; ++i){
@@ -66,6 +70,10 @@ connected_layer make_connected_layer(int batch, int inputs, int outputs, ACTIVAT
     }
 
 #ifdef GPU
+    l.forward_gpu = forward_connected_layer_gpu;
+    l.backward_gpu = backward_connected_layer_gpu;
+    l.update_gpu = update_connected_layer_gpu;
+
     l.weights_gpu = cuda_make_array(l.weights, outputs*inputs);
     l.biases_gpu = cuda_make_array(l.biases, outputs);
 

src/convolutional_layer.c

@@ -209,6 +209,9 @@ convolutional_layer make_convolutional_layer(int batch, int h, int w, int c, int
     l.output = calloc(l.batch*out_h * out_w * n, sizeof(float));
     l.delta  = calloc(l.batch*out_h * out_w * n, sizeof(float));
 
+    l.forward = forward_convolutional_layer;
+    l.backward = backward_convolutional_layer;
+    l.update = update_convolutional_layer;
     if(binary){
         l.binary_weights = calloc(c*n*size*size, sizeof(float));
         l.cweights = calloc(c*n*size*size, sizeof(char));
@@ -234,6 +237,10 @@ convolutional_layer make_convolutional_layer(int batch, int h, int w, int c, int
     }
 
 #ifdef GPU
+    l.forward_gpu = forward_convolutional_layer_gpu;
+    l.backward_gpu = backward_convolutional_layer_gpu;
+    l.update_gpu = update_convolutional_layer_gpu;
+
     if(gpu_index >= 0){
         l.weights_gpu = cuda_make_array(l.weights, c*n*size*size);
         l.weight_updates_gpu = cuda_make_array(l.weight_updates, c*n*size*size);

src/cost_layer.c

@@ -43,7 +43,13 @@ cost_layer make_cost_layer(int batch, int inputs, COST_TYPE cost_type, float sca
     l.delta = calloc(inputs*batch, sizeof(float));
     l.output = calloc(inputs*batch, sizeof(float));
     l.cost = calloc(1, sizeof(float));
+
+    l.forward = forward_cost_layer;
+    l.backward = backward_cost_layer;
     #ifdef GPU
+    l.forward_gpu = forward_cost_layer_gpu;
+    l.backward_gpu = backward_cost_layer_gpu;
+
     l.delta_gpu = cuda_make_array(l.output, inputs*batch);
     l.output_gpu = cuda_make_array(l.delta, inputs*batch);
     #endif

src/crnn_layer.c

@@ -64,7 +64,15 @@ layer make_crnn_layer(int batch, int h, int w, int c, int hidden_filters, int ou
     l.output = l.output_layer->output;
     l.delta = l.output_layer->delta;
 
+    l.forward = forward_crnn_layer;
+    l.backward = backward_crnn_layer;
+    l.update = update_crnn_layer;
+
 #ifdef GPU
+    l.forward_gpu = forward_crnn_layer_gpu;
+    l.backward_gpu = backward_crnn_layer_gpu;
+    l.update_gpu = update_crnn_layer_gpu;
+
     l.state_gpu = cuda_make_array(l.state, l.hidden*batch*(steps+1));
     l.output_gpu = l.output_layer->output_gpu;
     l.delta_gpu = l.output_layer->delta_gpu;

src/crop_layer.c

@@ -10,6 +10,9 @@ image get_crop_image(crop_layer l)
     return float_to_image(w,h,c,l.output);
 }
 
+void backward_crop_layer(const crop_layer l, network_state state){}
+void backward_crop_layer_gpu(const crop_layer l, network_state state){}
+
 crop_layer make_crop_layer(int batch, int h, int w, int c, int crop_height, int crop_width, int flip, float angle, float saturation, float exposure)
 {
     fprintf(stderr, "Crop Layer: %d x %d -> %d x %d x %d image\n", h,w,crop_height,crop_width,c);
@@ -30,7 +33,12 @@ crop_layer make_crop_layer(int batch, int h, int w, int c, int crop_height, int
     l.inputs = l.w * l.h * l.c;
     l.outputs = l.out_w * l.out_h * l.out_c;
     l.output = calloc(l.outputs*batch, sizeof(float));
+    l.forward = forward_crop_layer;
+    l.backward = backward_crop_layer;
+
     #ifdef GPU
+    l.forward_gpu = forward_crop_layer_gpu;
+    l.backward_gpu = backward_crop_layer_gpu;
     l.output_gpu = cuda_make_array(l.output, l.outputs*batch);
     l.rand_gpu   = cuda_make_array(0, l.batch*8);
     #endif

src/darknet.c

@@ -136,17 +136,6 @@ void partial(char *cfgfile, char *weightfile, char *outfile, int max)
     save_weights_upto(net, outfile, max);
 }
 
-void stacked(char *cfgfile, char *weightfile, char *outfile)
-{
-    gpu_index = -1;
-    network net = parse_network_cfg(cfgfile);
-    if(weightfile){
-        load_weights(&net, weightfile);
-    }
-    net.seen = 0;
-    save_weights_double(net, outfile);
-}
-
 #include "convolutional_layer.h"
 void rescale_net(char *cfgfile, char *weightfile, char *outfile)
 {
@@ -420,8 +409,6 @@ int main(int argc, char **argv)
         partial(argv[2], argv[3], argv[4], atoi(argv[5]));
     } else if (0 == strcmp(argv[1], "average")){
         average(argc, argv);
-    } else if (0 == strcmp(argv[1], "stacked")){
-        stacked(argv[2], argv[3], argv[4]);
     } else if (0 == strcmp(argv[1], "visualize")){
         visualize(argv[2], (argc > 3) ? argv[3] : 0);
     } else if (0 == strcmp(argv[1], "imtest")){