changing data loading

Joseph Redmon

2015-08-25 9d42f49a240136a8cd643cdc1f98230d4f22b05e

 src/coco.c

@@ -1,3 +1,5 @@
#include <stdio.h>

#include "network.h"
#include "detection_layer.h"
#include "cost_layer.h"
@@ -5,44 +7,40 @@
#include "parser.h"
#include "box.h"

#ifdef OPENCV
#include "opencv2/highgui/highgui_c.h"
#endif

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"};

void draw_coco(image im, float *box, int side, int objectness, char *label)
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};

void draw_coco(image im, float *pred, int side, char *label)
{
    int classes = 80;
    int elems = 4+classes+objectness;
    int classes = 81;
    int elems = 4+classes;
    int j;
    int r, c;

    for(r = 0; r < side; ++r){
        for(c = 0; c < side; ++c){
            j = (r*side + c) * elems;
            float scale = 1;
            if(objectness) scale = 1 - box[j++];
            int class = max_index(box+j, classes);
            if(scale * box[j+class] > 0.2){
                int width = box[j+class]*5 + 1;
                printf("%f %s\n", scale * box[j+class], coco_classes[class]);
            int class = max_index(pred+j, classes);
            if (class == 0) continue;
            if (pred[j+class] > 0.2){
                int width = pred[j+class]*5 + 1;
                printf("%f %s\n", pred[j+class], coco_classes[class-1]);
                float red = get_color(0,class,classes);
                float green = get_color(1,class,classes);
                float blue = get_color(2,class,classes);

                j += classes;
                float x = box[j+0];
                float y = box[j+1];
                x = (x+c)/side;
                y = (y+r)/side;
                float w = box[j+2]; //*maxwidth;
                float h = box[j+3]; //*maxheight;
                h = h*h;
                w = w*w;

                int left  = (x-w/2)*im.w;
                int right = (x+w/2)*im.w;
                int top   = (y-h/2)*im.h;
                int bot   = (y+h/2)*im.h;
                draw_box_width(im, left, top, right, bot, width, red, green, blue);
                box predict = {pred[j+0], pred[j+1], pred[j+2], pred[j+3]};
                box anchor = {(c+.5)/side, (r+.5)/side, .5, .5};
                box decode = decode_box(predict, anchor);
                
                draw_bbox(im, decode, width, red, green, blue);
            }
        }
    }
@@ -62,39 +60,47 @@
    if(weightfile){
        load_weights(&net, weightfile);
    }
    detection_layer layer = get_network_detection_layer(net);
    printf("Learning Rate: %g, Momentum: %g, Decay: %g\n", net.learning_rate, net.momentum, net.decay);
    int imgs = 128;
    int i = net.seen/imgs;
    data train, buffer;

    int classes = layer.classes;
    int background = layer.objectness;
    int side = sqrt(get_detection_layer_locations(layer));
    int classes = 81;
    int side = 7;

    char **paths;
    list *plist = get_paths(train_images);
    int N = plist->size;
    char **paths = (char **)list_to_array(plist);

    paths = (char **)list_to_array(plist);
    pthread_t load_thread = load_data_detection_thread(imgs, paths, plist->size, classes, net.w, net.h, side, side, background, &buffer);
    load_args args = {0};
    args.w = net.w;
    args.h = net.h;
    args.paths = paths;
    args.n = imgs;
    args.m = plist->size;
    args.classes = classes;
    args.num_boxes = side;
    args.d = &buffer;
    args.type = REGION_DATA;

    pthread_t load_thread = load_data_in_thread(args);
    clock_t time;
    while(i*imgs < N*120){
        i += 1;
        time=clock();
        pthread_join(load_thread, 0);
        train = buffer;
        load_thread = load_data_detection_thread(imgs, paths, plist->size, classes, net.w, net.h, side, side, background, &buffer);
        load_thread = load_data_in_thread(args);

        printf("Loaded: %lf seconds\n", sec(clock()-time));

        /*
           image im = float_to_image(net.w, net.h, 3, train.X.vals[114]);
           image copy = copy_image(im);
           draw_coco(copy, train.y.vals[114], 7, layer.objectness, "truth");
           cvWaitKey(0);
           free_image(copy);
         */
/*
        image im = float_to_image(net.w, net.h, 3, train.X.vals[114]);
        image copy = copy_image(im);
        draw_coco(copy, train.y.vals[114], 7, "truth");
        cvWaitKey(0);
        free_image(copy);
        */

        time=clock();
        float loss = train_network(net, train);
@@ -110,13 +116,12 @@
            save_weights(net, buff);
            return;
        }
        if(i%1000==0 || 1){
        if(i%1000==0){
            char buff[256];
            sprintf(buff, "%s/%s_%d.weights", backup_directory, base, i);
            save_weights(net, buff);
        }
        free_data(train);
        return;
    }
    char buff[256];
    sprintf(buff, "%s/%s_final.weights", backup_directory, base);
@@ -145,7 +150,7 @@
    }
}

void print_cocos(FILE **fps, char *id, box *boxes, float **probs, int num_boxes, int classes, int w, int h)
void print_cocos(FILE *fp, int image_id, box *boxes, float **probs, int num_boxes, int classes, int w, int h)
{
    int i, j;
    for(i = 0; i < num_boxes*num_boxes; ++i){
@@ -159,13 +164,23 @@
        if (xmax > w) xmax = w;
        if (ymax > h) ymax = h;

        float bx = xmin;
        float by = ymin;
        float bw = xmax - xmin;
        float bh = ymax - ymin;

        for(j = 0; j < classes; ++j){
            if (probs[i][j]) fprintf(fps[j], "%s %f %f %f %f %f\n", id, probs[i][j],
                    xmin, ymin, xmax, ymax);
            if (probs[i][j]) fprintf(fp, "{\"image_id\":%d, \"category_id\":%d, \"bbox\":[%f, %f, %f, %f], \"score\":%f},\n", image_id, coco_ids[j], bx, by, bw, bh, probs[i][j]);
        }
    }
}

int get_coco_image_id(char *filename)
{
    char *p = strrchr(filename, '_');
    return atoi(p+1);
}

void validate_coco(char *cfgfile, char *weightfile)
{
    network net = parse_network_cfg(cfgfile);
@@ -177,8 +192,8 @@
    fprintf(stderr, "Learning Rate: %g, Momentum: %g, Decay: %g\n", net.learning_rate, net.momentum, net.decay);
    srand(time(0));

    char *base = "results/comp4_det_test_";
    list *plist = get_paths("data/voc.2012test.list");
    char *base = "/home/pjreddie/backup/";
    list *plist = get_paths("data/coco_val_5k.list");
    char **paths = (char **)list_to_array(plist);

    int classes = layer.classes;
@@ -187,12 +202,11 @@
    int num_boxes = sqrt(get_detection_layer_locations(layer));

    int j;
    FILE **fps = calloc(classes, sizeof(FILE *));
    for(j = 0; j < classes; ++j){
        char buff[1024];
        snprintf(buff, 1024, "%s%s.txt", base, coco_classes[j]);
        fps[j] = fopen(buff, "w");
    }
    char buff[1024];
    snprintf(buff, 1024, "%s/coco_results.json", base);
    FILE *fp = fopen(buff, "w");
    fprintf(fp, "[\n");

    box *boxes = calloc(num_boxes*num_boxes, sizeof(box));
    float **probs = calloc(num_boxes*num_boxes, sizeof(float *));
    for(j = 0; j < num_boxes*num_boxes; ++j) probs[j] = calloc(classes, sizeof(float *));
@@ -201,10 +215,15 @@
    int i=0;
    int t;

    float thresh = .001;
    float thresh = .01;
    int nms = 1;
    float iou_thresh = .5;

    load_args args = {0};
    args.w = net.w;
    args.h = net.h;
    args.type = IMAGE_DATA;

    int nthreads = 8;
    image *val = calloc(nthreads, sizeof(image));
    image *val_resized = calloc(nthreads, sizeof(image));
@@ -212,7 +231,10 @@
    image *buf_resized = calloc(nthreads, sizeof(image));
    pthread_t *thr = calloc(nthreads, sizeof(pthread_t));
    for(t = 0; t < nthreads; ++t){
        thr[t] = load_image_thread(paths[i+t], &buf[t], &buf_resized[t], net.w, net.h);
        args.path = paths[i+t];
        args.im = &buf[t];
        args.resized = &buf_resized[t];
        thr[t] = load_data_in_thread(args);
    }
    time_t start = time(0);
    for(i = nthreads; i < m+nthreads; i += nthreads){
@@ -223,23 +245,28 @@
            val_resized[t] = buf_resized[t];
        }
        for(t = 0; t < nthreads && i+t < m; ++t){
            thr[t] = load_image_thread(paths[i+t], &buf[t], &buf_resized[t], net.w, net.h);
            args.path = paths[i+t];
            args.im = &buf[t];
            args.resized = &buf_resized[t];
            thr[t] = load_data_in_thread(args);
        }
        for(t = 0; t < nthreads && i+t-nthreads < m; ++t){
            char *path = paths[i+t-nthreads];
            char *id = basecfg(path);
            int image_id = get_coco_image_id(path);
            float *X = val_resized[t].data;
            float *predictions = network_predict(net, X);
            int w = val[t].w;
            int h = val[t].h;
            convert_cocos(predictions, classes, objectness, background, num_boxes, w, h, thresh, probs, boxes);
            if (nms) do_nms(boxes, probs, num_boxes, classes, iou_thresh);
            print_cocos(fps, id, boxes, probs, num_boxes, classes, w, h);
            free(id);
            print_cocos(fp, image_id, boxes, probs, num_boxes, classes, w, h);
            free_image(val[t]);
            free_image(val_resized[t]);
        }
    }
    fseek(fp, -2, SEEK_CUR); 
    fprintf(fp, "\n]\n");
    fclose(fp);
    fprintf(stderr, "Total Detection Time: %f Seconds\n", (double)(time(0) - start));
}

@@ -250,7 +277,6 @@
    if(weightfile){
        load_weights(&net, weightfile);
    }
    detection_layer layer = get_network_detection_layer(net);
    set_batch_network(&net, 1);
    srand(2222222);
    clock_t time;
@@ -270,7 +296,7 @@
        time=clock();
        float *predictions = network_predict(net, X);
        printf("%s: Predicted in %f seconds.\n", input, sec(clock()-time));
        draw_coco(im, predictions, 7, layer.objectness, "predictions");
        draw_coco(im, predictions, 7, "predictions");
        free_image(im);
        free_image(sized);
#ifdef OPENCV