| | |
| | | #include "softmax_layer.h" |
| | | #include "dropout_layer.h" |
| | | #include "detection_layer.h" |
| | | #include "region_layer.h" |
| | | #include "avgpool_layer.h" |
| | | #include "local_layer.h" |
| | | #include "route_layer.h" |
| | | #include "shortcut_layer.h" |
| | | #include "list.h" |
| | | #include "option_list.h" |
| | | #include "utils.h" |
| | |
| | | |
| | | int is_network(section *s); |
| | | int is_convolutional(section *s); |
| | | int is_local(section *s); |
| | | int is_deconvolutional(section *s); |
| | | int is_connected(section *s); |
| | | int is_maxpool(section *s); |
| | |
| | | int is_softmax(section *s); |
| | | int is_normalization(section *s); |
| | | int is_crop(section *s); |
| | | int is_shortcut(section *s); |
| | | int is_cost(section *s); |
| | | int is_detection(section *s); |
| | | int is_region(section *s); |
| | | int is_route(section *s); |
| | | list *read_cfg(char *filename); |
| | | |
| | |
| | | int h; |
| | | int w; |
| | | int c; |
| | | int index; |
| | | } size_params; |
| | | |
| | | deconvolutional_layer parse_deconvolutional(list *options, size_params params) |
| | |
| | | return layer; |
| | | } |
| | | |
| | | local_layer parse_local(list *options, size_params params) |
| | | { |
| | | int n = option_find_int(options, "filters",1); |
| | | int size = option_find_int(options, "size",1); |
| | | int stride = option_find_int(options, "stride",1); |
| | | int pad = option_find_int(options, "pad",0); |
| | | char *activation_s = option_find_str(options, "activation", "logistic"); |
| | | ACTIVATION activation = get_activation(activation_s); |
| | | |
| | | int batch,h,w,c; |
| | | h = params.h; |
| | | w = params.w; |
| | | c = params.c; |
| | | batch=params.batch; |
| | | if(!(h && w && c)) error("Layer before local layer must output image."); |
| | | |
| | | local_layer layer = make_local_layer(batch,h,w,c,n,size,stride,pad,activation); |
| | | |
| | | return layer; |
| | | } |
| | | |
| | | convolutional_layer parse_convolutional(list *options, size_params params) |
| | | { |
| | | int n = option_find_int(options, "filters",1); |
| | |
| | | c = params.c; |
| | | batch=params.batch; |
| | | if(!(h && w && c)) error("Layer before convolutional layer must output image."); |
| | | int batch_normalize = option_find_int_quiet(options, "batch_normalize", 0); |
| | | |
| | | convolutional_layer layer = make_convolutional_layer(batch,h,w,c,n,size,stride,pad,activation); |
| | | convolutional_layer layer = make_convolutional_layer(batch,h,w,c,n,size,stride,pad,activation, batch_normalize); |
| | | layer.flipped = option_find_int_quiet(options, "flipped", 0); |
| | | |
| | | char *weights = option_find_str(options, "weights", 0); |
| | | char *biases = option_find_str(options, "biases", 0); |
| | |
| | | int coords = option_find_int(options, "coords", 1); |
| | | int classes = option_find_int(options, "classes", 1); |
| | | int rescore = option_find_int(options, "rescore", 0); |
| | | int joint = option_find_int(options, "joint", 0); |
| | | int objectness = option_find_int(options, "objectness", 0); |
| | | int background = 0; |
| | | detection_layer layer = make_detection_layer(params.batch, params.inputs, classes, coords, joint, rescore, background, objectness); |
| | | return layer; |
| | | } |
| | | |
| | | region_layer parse_region(list *options, size_params params) |
| | | { |
| | | int coords = option_find_int(options, "coords", 1); |
| | | int classes = option_find_int(options, "classes", 1); |
| | | int rescore = option_find_int(options, "rescore", 0); |
| | | int num = option_find_int(options, "num", 1); |
| | | int side = option_find_int(options, "side", 7); |
| | | region_layer layer = make_region_layer(params.batch, params.inputs, num, side, classes, coords, rescore); |
| | | detection_layer layer = make_detection_layer(params.batch, params.inputs, num, side, classes, coords, rescore); |
| | | |
| | | layer.softmax = option_find_int(options, "softmax", 0); |
| | | layer.sqrt = option_find_int(options, "sqrt", 0); |
| | | |
| | | layer.coord_scale = option_find_float(options, "coord_scale", 1); |
| | | layer.forced = option_find_int(options, "forced", 0); |
| | | layer.object_scale = option_find_float(options, "object_scale", 1); |
| | | 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); |
| | | return layer; |
| | | } |
| | | |
| | |
| | | int noadjust = option_find_int_quiet(options, "noadjust",0); |
| | | |
| | | crop_layer l = make_crop_layer(batch,h,w,c,crop_height,crop_width,flip, angle, saturation, exposure); |
| | | l.shift = option_find_float(options, "shift", 0); |
| | | l.noadjust = noadjust; |
| | | return l; |
| | | } |
| | |
| | | return l; |
| | | } |
| | | |
| | | layer parse_shortcut(list *options, size_params params, network net) |
| | | { |
| | | char *l = option_find(options, "from"); |
| | | int index = atoi(l); |
| | | if(index < 0) index = params.index + index; |
| | | |
| | | int batch = params.batch; |
| | | layer from = net.layers[index]; |
| | | |
| | | layer s = make_shortcut_layer(batch, index, params.w, params.h, params.c, from.out_w, from.out_h, from.out_c); |
| | | return s; |
| | | } |
| | | |
| | | |
| | | route_layer parse_route(list *options, size_params params, network net) |
| | | { |
| | | char *l = option_find(options, "layers"); |
| | |
| | | for(i = 0; i < n; ++i){ |
| | | int index = atoi(l); |
| | | l = strchr(l, ',')+1; |
| | | if(index < 0) index = params.index + index; |
| | | layers[i] = index; |
| | | sizes[i] = net.layers[index].outputs; |
| | | } |
| | | int batch = params.batch; |
| | | |
| | | route_layer layer = make_route_layer(batch, n, layers, sizes); |
| | | |
| | | |
| | | convolutional_layer first = net.layers[layers[0]]; |
| | | layer.out_w = first.out_w; |
| | | layer.out_h = first.out_h; |
| | |
| | | if (strcmp(s, "constant")==0) return CONSTANT; |
| | | if (strcmp(s, "step")==0) return STEP; |
| | | if (strcmp(s, "exp")==0) return EXP; |
| | | if (strcmp(s, "sigmoid")==0) return SIG; |
| | | if (strcmp(s, "steps")==0) return STEPS; |
| | | fprintf(stderr, "Couldn't find policy %s, going with constant\n", s); |
| | | return CONSTANT; |
| | | } |
| | |
| | | net->policy = get_policy(policy_s); |
| | | if(net->policy == STEP){ |
| | | net->step = option_find_int(options, "step", 1); |
| | | net->gamma = option_find_float(options, "gamma", 1); |
| | | net->scale = option_find_float(options, "scale", 1); |
| | | } else if (net->policy == STEPS){ |
| | | char *l = option_find(options, "steps"); |
| | | char *p = option_find(options, "scales"); |
| | | if(!l || !p) error("STEPS policy must have steps and scales in cfg file"); |
| | | |
| | | int len = strlen(l); |
| | | int n = 1; |
| | | int i; |
| | | for(i = 0; i < len; ++i){ |
| | | if (l[i] == ',') ++n; |
| | | } |
| | | int *steps = calloc(n, sizeof(int)); |
| | | float *scales = calloc(n, sizeof(float)); |
| | | for(i = 0; i < n; ++i){ |
| | | int step = atoi(l); |
| | | float scale = atof(p); |
| | | l = strchr(l, ',')+1; |
| | | p = strchr(p, ',')+1; |
| | | steps[i] = step; |
| | | scales[i] = scale; |
| | | } |
| | | net->scales = scales; |
| | | net->steps = steps; |
| | | net->num_steps = n; |
| | | } else if (net->policy == EXP){ |
| | | net->gamma = option_find_float(options, "gamma", 1); |
| | | } else if (net->policy == SIG){ |
| | | net->gamma = option_find_float(options, "gamma", 1); |
| | | net->step = option_find_int(options, "step", 1); |
| | | } else if (net->policy == POLY){ |
| | | net->power = option_find_float(options, "power", 1); |
| | | } |
| | |
| | | int count = 0; |
| | | free_section(s); |
| | | while(n){ |
| | | params.index = count; |
| | | fprintf(stderr, "%d: ", count); |
| | | s = (section *)n->val; |
| | | options = s->options; |
| | | layer l = {0}; |
| | | if(is_convolutional(s)){ |
| | | l = parse_convolutional(options, params); |
| | | }else if(is_local(s)){ |
| | | l = parse_local(options, params); |
| | | }else if(is_deconvolutional(s)){ |
| | | l = parse_deconvolutional(options, params); |
| | | }else if(is_connected(s)){ |
| | |
| | | l = parse_cost(options, params); |
| | | }else if(is_detection(s)){ |
| | | l = parse_detection(options, params); |
| | | }else if(is_region(s)){ |
| | | l = parse_region(options, params); |
| | | }else if(is_softmax(s)){ |
| | | l = parse_softmax(options, params); |
| | | }else if(is_normalization(s)){ |
| | |
| | | l = parse_avgpool(options, params); |
| | | }else if(is_route(s)){ |
| | | l = parse_route(options, params, net); |
| | | }else if(is_shortcut(s)){ |
| | | l = parse_shortcut(options, params, net); |
| | | }else if(is_dropout(s)){ |
| | | l = parse_dropout(options, params); |
| | | l.output = net.layers[count-1].output; |
| | | l.delta = net.layers[count-1].delta; |
| | | #ifdef GPU |
| | | #ifdef GPU |
| | | l.output_gpu = net.layers[count-1].output_gpu; |
| | | l.delta_gpu = net.layers[count-1].delta_gpu; |
| | | #endif |
| | | #endif |
| | | }else{ |
| | | fprintf(stderr, "Type not recognized: %s\n", s->type); |
| | | } |
| | | l.dontload = option_find_int_quiet(options, "dontload", 0); |
| | | l.dontloadscales = option_find_int_quiet(options, "dontloadscales", 0); |
| | | option_unused(options); |
| | | net.layers[count] = l; |
| | | free_section(s); |
| | | n = n->next; |
| | | ++count; |
| | | if(n){ |
| | | params.h = l.out_h; |
| | | params.w = l.out_w; |
| | | params.c = l.out_c; |
| | | params.inputs = l.outputs; |
| | | } |
| | | ++count; |
| | | } |
| | | free_list(sections); |
| | | net.outputs = get_network_output_size(net); |
| | |
| | | return net; |
| | | } |
| | | |
| | | int is_shortcut(section *s) |
| | | { |
| | | return (strcmp(s->type, "[shortcut]")==0); |
| | | } |
| | | int is_crop(section *s) |
| | | { |
| | | return (strcmp(s->type, "[crop]")==0); |
| | |
| | | { |
| | | return (strcmp(s->type, "[detection]")==0); |
| | | } |
| | | int is_region(section *s) |
| | | int is_local(section *s) |
| | | { |
| | | return (strcmp(s->type, "[region]")==0); |
| | | return (strcmp(s->type, "[local]")==0); |
| | | } |
| | | int is_deconvolutional(section *s) |
| | | { |
| | |
| | | return (strcmp(s->type, "[route]")==0); |
| | | } |
| | | |
| | | int read_option(char *s, list *options) |
| | | { |
| | | size_t i; |
| | | size_t len = strlen(s); |
| | | char *val = 0; |
| | | for(i = 0; i < len; ++i){ |
| | | if(s[i] == '='){ |
| | | s[i] = '\0'; |
| | | val = s+i+1; |
| | | break; |
| | | } |
| | | } |
| | | if(i == len-1) return 0; |
| | | char *key = s; |
| | | option_insert(options, key, val); |
| | | return 1; |
| | | } |
| | | |
| | | list *read_cfg(char *filename) |
| | | { |
| | | FILE *file = fopen(filename, "r"); |
| | |
| | | FILE *fp = fopen(filename, "w"); |
| | | if(!fp) file_error(filename); |
| | | |
| | | fwrite(&net.learning_rate, sizeof(float), 1, fp); |
| | | fwrite(&net.momentum, sizeof(float), 1, fp); |
| | | fwrite(&net.decay, sizeof(float), 1, fp); |
| | | int major = 0; |
| | | int minor = 1; |
| | | int revision = 0; |
| | | fwrite(&major, sizeof(int), 1, fp); |
| | | fwrite(&minor, sizeof(int), 1, fp); |
| | | fwrite(&revision, sizeof(int), 1, fp); |
| | | fwrite(net.seen, sizeof(int), 1, fp); |
| | | |
| | | int i; |
| | |
| | | #endif |
| | | int num = l.n*l.c*l.size*l.size; |
| | | fwrite(l.biases, sizeof(float), l.n, fp); |
| | | fwrite(l.filters, sizeof(float), num, fp); |
| | | } |
| | | if(l.type == DECONVOLUTIONAL){ |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | pull_deconvolutional_layer(l); |
| | | if (l.batch_normalize){ |
| | | fwrite(l.scales, sizeof(float), l.n, fp); |
| | | fwrite(l.rolling_mean, sizeof(float), l.n, fp); |
| | | fwrite(l.rolling_variance, sizeof(float), l.n, fp); |
| | | } |
| | | #endif |
| | | int num = l.n*l.c*l.size*l.size; |
| | | fwrite(l.biases, sizeof(float), l.n, fp); |
| | | fwrite(l.filters, sizeof(float), num, fp); |
| | | } |
| | | if(l.type == CONNECTED){ |
| | | } if(l.type == CONNECTED){ |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | pull_connected_layer(l); |
| | |
| | | #endif |
| | | fwrite(l.biases, sizeof(float), l.outputs, fp); |
| | | fwrite(l.weights, sizeof(float), l.outputs*l.inputs, fp); |
| | | } if(l.type == LOCAL){ |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | pull_local_layer(l); |
| | | } |
| | | #endif |
| | | int locations = l.out_w*l.out_h; |
| | | int size = l.size*l.size*l.c*l.n*locations; |
| | | fwrite(l.biases, sizeof(float), l.outputs, fp); |
| | | fwrite(l.filters, sizeof(float), size, fp); |
| | | } |
| | | } |
| | | fclose(fp); |
| | |
| | | save_weights_upto(net, filename, net.n); |
| | | } |
| | | |
| | | void transpose_matrix(float *a, int rows, int cols) |
| | | { |
| | | float *transpose = calloc(rows*cols, sizeof(float)); |
| | | int x, y; |
| | | for(x = 0; x < rows; ++x){ |
| | | for(y = 0; y < cols; ++y){ |
| | | transpose[y*rows + x] = a[x*cols + y]; |
| | | } |
| | | } |
| | | memcpy(a, transpose, rows*cols*sizeof(float)); |
| | | free(transpose); |
| | | } |
| | | |
| | | void load_weights_upto(network *net, char *filename, int cutoff) |
| | | { |
| | | fprintf(stderr, "Loading weights from %s...", filename); |
| | |
| | | FILE *fp = fopen(filename, "r"); |
| | | if(!fp) file_error(filename); |
| | | |
| | | float garbage; |
| | | fread(&garbage, sizeof(float), 1, fp); |
| | | fread(&garbage, sizeof(float), 1, fp); |
| | | fread(&garbage, sizeof(float), 1, fp); |
| | | int major; |
| | | int minor; |
| | | int revision; |
| | | fread(&major, sizeof(int), 1, fp); |
| | | fread(&minor, sizeof(int), 1, fp); |
| | | fread(&revision, sizeof(int), 1, fp); |
| | | fread(net->seen, sizeof(int), 1, fp); |
| | | |
| | | int i; |
| | |
| | | if(l.type == CONVOLUTIONAL){ |
| | | int num = l.n*l.c*l.size*l.size; |
| | | fread(l.biases, sizeof(float), l.n, fp); |
| | | if (l.batch_normalize && (!l.dontloadscales)){ |
| | | fread(l.scales, sizeof(float), l.n, fp); |
| | | fread(l.rolling_mean, sizeof(float), l.n, fp); |
| | | fread(l.rolling_variance, sizeof(float), l.n, fp); |
| | | } |
| | | fread(l.filters, sizeof(float), num, fp); |
| | | if (l.flipped) { |
| | | transpose_matrix(l.filters, l.c*l.size*l.size, l.n); |
| | | } |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | push_convolutional_layer(l); |
| | |
| | | if(l.type == CONNECTED){ |
| | | fread(l.biases, sizeof(float), l.outputs, fp); |
| | | fread(l.weights, sizeof(float), l.outputs*l.inputs, fp); |
| | | if(major > 1000 || minor > 1000){ |
| | | transpose_matrix(l.weights, l.inputs, l.outputs); |
| | | } |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | push_connected_layer(l); |
| | | } |
| | | #endif |
| | | } |
| | | if(l.type == LOCAL){ |
| | | int locations = l.out_w*l.out_h; |
| | | int size = l.size*l.size*l.c*l.n*locations; |
| | | fread(l.biases, sizeof(float), l.outputs, fp); |
| | | fread(l.filters, sizeof(float), size, fp); |
| | | #ifdef GPU |
| | | if(gpu_index >= 0){ |
| | | push_local_layer(l); |
| | | } |
| | | #endif |
| | | } |
| | | } |
| | | fprintf(stderr, "Done!\n"); |
| | | fclose(fp); |