YOLO原始碼分析之data.c
阿新 • • 發佈：2019-01-21
darknet裡樣本的儲存是以如下形式進行排列的
flag | class_info | box_info
flag為0或者1，表示有沒有物體
class_info表示類別資訊，其長度是num_class
box_info表示標註資訊，其長度是5(x, y, w, h, o)
d.X.val指向實際的影象資料，其經過深度網路後，變成detection_layer或者region_layer裡的l.output
d.y.val指向實際的標註資訊，其範圍已經轉換到[0,1]，其在detection_layer或者region_layer裡是state.truth
#include "data.h" #include "utils.h" #include "image.h" #include "cuda.h" #include <stdio.h> #include <stdlib.h> #include <string.h> pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER; list *get_paths(char *filename) { char *path; FILE *file = fopen(filename, "r"); if(!file) file_error(filename); list *lines = make_list(); while((path=fgetl(file))){ list_insert(lines, path); } fclose(file); return lines; } /* char **get_random_paths_indexes(char **paths, int n, int m, int *indexes) { char **random_paths = calloc(n, sizeof(char*)); int i; pthread_mutex_lock(&mutex); for(i = 0; i < n; ++i){ int index = rand()%m; indexes[i] = index; random_paths[i] = paths[index]; if(i == 0) printf("%s\n", paths[index]); } pthread_mutex_unlock(&mutex); return random_paths; } */ char **get_random_paths(char **paths, int n, int m) { char **random_paths = calloc(n, sizeof(char*)); int i; pthread_mutex_lock(&mutex); for(i = 0; i < n; ++i){ int index = rand()%m; random_paths[i] = paths[index]; //if(i == 0) printf("%s\n", paths[index]); } pthread_mutex_unlock(&mutex); return random_paths; } char **find_replace_paths(char **paths, int n, char *find, char *replace) { char **replace_paths = calloc(n, sizeof(char*)); int i; for(i = 0; i < n; ++i){ char replaced[4096]; find_replace(paths[i], find, replace, replaced); replace_paths[i] = copy_string(replaced); } return replace_paths; } matrix load_image_paths_gray(char **paths, int n, int w, int h) { int i; matrix X; X.rows = n; X.vals = calloc(X.rows, sizeof(float*)); X.cols = 0; for(i = 0; i < n; ++i){ image im = load_image(paths[i], w, h, 3); image gray = grayscale_image(im); free_image(im); im = gray; X.vals[i] = im.data; X.cols = im.h*im.w*im.c; } return X; } matrix load_image_paths(char **paths, int n, int w, int h) { int i; matrix X; X.rows = n; X.vals = calloc(X.rows, sizeof(float*)); X.cols = 0; for(i = 0; i < n; ++i){ image im = load_image_color(paths[i], w, h); X.vals[i] = im.data; X.cols = im.h*im.w*im.c; } return X; } matrix load_image_augment_paths(char **paths, int n, int min, int max, int size, float angle, float aspect, float hue, float saturation, float exposure) { int i; matrix X; X.rows = n; X.vals = calloc(X.rows, sizeof(float*)); X.cols = 0; for(i = 0; i < n; ++i){ image im = load_image_color(paths[i], 0, 0); image crop = random_augment_image(im, angle, aspect, min, max, size); int flip = rand()%2; if (flip) flip_image(crop); random_distort_image(crop, hue, saturation, exposure); /* show_image(im, "orig"); show_image(crop, "crop"); cvWaitKey(0); */ free_image(im); X.vals[i] = crop.data; X.cols = crop.h*crop.w*crop.c; } return X; } box_label *read_boxes(char *filename, int *n) { box_label *boxes = calloc(1, sizeof(box_label)); FILE *file = fopen(filename, "r"); if(!file) file_error(filename); float x, y, h, w; int id; int count = 0; while(fscanf(file, "%d %f %f %f %f", &id, &x, &y, &w, &h) == 5){ boxes = realloc(boxes, (count+1)*sizeof(box_label)); boxes[count].id = id; boxes[count].x = x; boxes[count].y = y; boxes[count].h = h; boxes[count].w = w; boxes[count].left = x - w/2; boxes[count].right = x + w/2; boxes[count].top = y - h/2; boxes[count].bottom = y + h/2; ++count; } fclose(file); *n = count; return boxes; } void randomize_boxes(box_label *b, int n) { int i; for(i = 0; i < n; ++i){ box_label swap = b[i]; int index = rand()%n; b[i] = b[index]; b[index] = swap; } } void correct_boxes(box_label *boxes, int n, float dx, float dy, float sx, float sy, int flip) { int i; for(i = 0; i < n; ++i){ if(boxes[i].x == 0 && boxes[i].y == 0) { boxes[i].x = 999999; boxes[i].y = 999999; boxes[i].w = 999999; boxes[i].h = 999999; continue; } boxes[i].left = boxes[i].left * sx - dx; boxes[i].right = boxes[i].right * sx - dx; boxes[i].top = boxes[i].top * sy - dy; boxes[i].bottom = boxes[i].bottom* sy - dy; if(flip){ float swap = boxes[i].left; boxes[i].left = 1. - boxes[i].right; boxes[i].right = 1. - swap; } boxes[i].left = constrain(0, 1, boxes[i].left); boxes[i].right = constrain(0, 1, boxes[i].right); boxes[i].top = constrain(0, 1, boxes[i].top); boxes[i].bottom = constrain(0, 1, boxes[i].bottom); boxes[i].x = (boxes[i].left+boxes[i].right)/2; boxes[i].y = (boxes[i].top+boxes[i].bottom)/2; boxes[i].w = (boxes[i].right - boxes[i].left); boxes[i].h = (boxes[i].bottom - boxes[i].top); boxes[i].w = constrain(0, 1, boxes[i].w); boxes[i].h = constrain(0, 1, boxes[i].h); } } void fill_truth_swag(char *path, float *truth, int classes, int flip, float dx, float dy, float sx, float sy) { char labelpath[4096]; find_replace(path, "images", "labels", labelpath); find_replace(labelpath, "JPEGImages", "labels", labelpath); find_replace(labelpath, ".jpg", ".txt", labelpath); find_replace(labelpath, ".JPG", ".txt", labelpath); find_replace(labelpath, ".JPEG", ".txt", labelpath); int count = 0; box_label *boxes = read_boxes(labelpath, &count); randomize_boxes(boxes, count); correct_boxes(boxes, count, dx, dy, sx, sy, flip); float x,y,w,h; int id; int i; for (i = 0; i < count && i < 30; ++i) { x = boxes[i].x; y = boxes[i].y; w = boxes[i].w; h = boxes[i].h; id = boxes[i].id; if (w < .0 || h < .0) continue; int index = (4+classes) * i; truth[index++] = x; truth[index++] = y; truth[index++] = w; truth[index++] = h; if (id < classes) truth[index+id] = 1; } free(boxes); } void fill_truth_region(char *path, float *truth, int classes, int num_boxes, int flip, float dx, float dy, float sx, float sy) { char labelpath[4096]; find_replace(path, "images", "labels", labelpath); find_replace(labelpath, "JPEGImages", "labels", labelpath); find_replace(labelpath, ".jpg", ".txt", labelpath); find_replace(labelpath, ".png", ".txt", labelpath); find_replace(labelpath, ".JPG", ".txt", labelpath); find_replace(labelpath, ".JPEG", ".txt", labelpath); int count = 0; box_label *boxes = read_boxes(labelpath, &count); randomize_boxes(boxes, count); correct_boxes(boxes, count, dx, dy, sx, sy, flip); float x,y,w,h; int id; int i; for (i = 0; i < count; ++i) { x = boxes[i].x; y = boxes[i].y; w = boxes[i].w; h = boxes[i].h; id = boxes[i].id; if (w < .01 || h < .01) continue; int col = (int)(x*num_boxes); int row = (int)(y*num_boxes); x = x*num_boxes - col; y = y*num_boxes - row; int index = (col+row*num_boxes)*(5+classes); if (truth[index]) continue;
// 從這裡可以看出truth是一個grid*grid*(5+classes)的陣列，其中5代表著// [0] is_obj // [1] x// [2] y// [3] w// [4] h
// 然後truth的[5]--[5+classes]就是id了memcpy(y+j*d.y.cols, d.y.vals[index], d.y.cols*sizeof(float)); }}void smooth_data(data d){ int i, j; float scale = 1. / d.y.cols; float eps = .1; for(i = 0; i < d.y.rows; ++i){ for(j = 0; j < d.y.cols; ++j){ d.y.vals[i][j] = eps * scale + (1-eps) * d.y.vals[i][j]; } }}data load_all_cifar10(){ data d = {0}; d.shallow = 0; int i,j,b; matrix X = make_matrix(50000, 3072); matrix y = make_matrix(50000, 10); d.X = X; d.y = y; for(b = 0; b < 5; ++b){ char buff[256]; sprintf(buff, "data/cifar/cifar-10-batches-bin/data_batch_%d.bin", b+1); FILE *fp = fopen(buff, "rb"); if(!fp) file_error(buff); for(i = 0; i < 10000; ++i){ unsigned char bytes[3073]; fread(bytes, 1, 3073, fp); int class = bytes[0]; y.vals[i+b*10000][class] = 1; for(j = 0; j < X.cols; ++j){ X.vals[i+b*10000][j] = (double)bytes[j+1]; } } fclose(fp); } //normalize_data_rows(d); //translate_data_rows(d, -128); scale_data_rows(d, 1./255); smooth_data(d); return d;}data load_go(char *filename){ FILE *fp = fopen(filename, "rb"); matrix X = make_matrix(3363059, 361); matrix y = make_matrix(3363059, 361); int row, col; if(!fp) file_error(filename); char *label; int count = 0; while((label = fgetl(fp))){ int i; if(count == X.rows){ X = resize_matrix(X, count*2); y = resize_matrix(y, count*2); } sscanf(label, "%d %d", &row, &col); char *board = fgetl(fp); int index = row*19 + col; y.vals[count][index] = 1; for(i = 0; i < 19*19; ++i){ float val = 0; if(board[i] == '1') val = 1; else if(board[i] == '2') val = -1; X.vals[count][i] = val; } ++count; free(label); free(board); } X = resize_matrix(X, count); y = resize_matrix(y, count); data d = {0}; d.shallow = 0; d.X = X; d.y = y; fclose(fp); return d;}void randomize_data(data d){ int i; for(i = d.X.rows-1; i > 0; --i){ int index = rand()%i; float *swap = d.X.vals[index]; d.X.vals[index] = d.X.vals[i]; d.X.vals[i] = swap; swap = d.y.vals[index]; d.y.vals[index] = d.y.vals[i]; d.y.vals[i] = swap; }}void scale_data_rows(data d, float s){ int i; for(i = 0; i < d.X.rows; ++i){ scale_array(d.X.vals[i], d.X.cols, s); }}void translate_data_rows(data d, float s){ int i; for(i = 0; i < d.X.rows; ++i){ translate_array(d.X.vals[i], d.X.cols, s); }}void normalize_data_rows(data d){ int i; for(i = 0; i < d.X.rows; ++i){ normalize_array(d.X.vals[i], d.X.cols); }}data get_data_part(data d, int part, int total){ data p = {0}; p.shallow = 1; p.X.rows = d.X.rows * (part + 1) / total - d.X.rows * part / total; p.y.rows = d.y.rows * (part + 1) / total - d.y.rows * part / total; p.X.cols = d.X.cols; p.y.cols = d.y.cols; p.X.vals = d.X.vals + d.X.rows * part / total; p.y.vals = d.y.vals + d.y.rows * part / total; return p;}data get_random_data(data d, int num){ data r = {0}; r.shallow = 1; r.X.rows = num; r.y.rows = num; r.X.cols = d.X.cols; r.y.cols = d.y.cols; r.X.vals = calloc(num, sizeof(float *)); r.y.vals = calloc(num, sizeof(float *)); int i; for(i = 0; i < num; ++i){ int index = rand()%d.X.rows; r.X.vals[i] = d.X.vals[index]; r.y.vals[i] = d.y.vals[index]; } return r;}data *split_data(data d, int part, int total){ data *split = calloc(2, sizeof(data)); int i; int start = part*d.X.rows/total; int end = (part+1)*d.X.rows/total; data train; data test; train.shallow = test.shallow = 1; test.X.rows = test.y.rows = end-start; train.X.rows = train.y.rows = d.X.rows - (end-start); train.X.cols = test.X.cols = d.X.cols; train.y.cols = test.y.cols = d.y.cols; train.X.vals = calloc(train.X.rows, sizeof(float*)); test.X.vals = calloc(test.X.rows, sizeof(float*)); train.y.vals = calloc(train.y.rows, sizeof(float*)); test.y.vals = calloc(test.y.rows, sizeof(float*)); for(i = 0; i < start; ++i){ train.X.vals[i] = d.X.vals[i]; train.y.vals[i] = d.y.vals[i]; } for(i = start; i < end; ++i){ test.X.vals[i-start] = d.X.vals[i]; test.y.vals[i-start] = d.y.vals[i]; } for(i = end; i < d.X.rows; ++i){ train.X.vals[i-(end-start)] = d.X.vals[i]; train.y.vals[i-(end-start)] = d.y.vals[i]; } split[0] = train; split[1] = test; return split;}// batch的索引地址，二級指標指向實際的影象資料，二級指標的size即為h*w*c