mirror/darknet
1
0
Fork 0
mirror of https://github.com/pjreddie/darknet.git synced 2023-08-10 21:13:14 +03:00
Code Issues Projects Releases Wiki Activity

Training on VOC

Browse Source
This commit is contained in:
Joseph Redmon
2014-02-14 10:26:31 -08:00
parent f7a17f82eb
commit 118bdd6f62
26 changed files with 501 additions and 240 deletions
Download Patch File Download Diff File Expand all files Collapse all files

187
src/parser.c
View File

@ -23,6 +23,130 @@ int is_maxpool(section *s);
int is_softmax(section *s);
list *read_cfg(char *filename);
void free_section(section *s)
{
free(s->type);
node *n = s->options->front;
while(n){
kvp *pair = (kvp *)n->val;
free(pair->key);
free(pair);
node *next = n->next;
free(n);
n = next;
}
free(s->options);
free(s);
}
convolutional_layer *parse_convolutional(list *options, network net, int count)
{
int i;
int h,w,c;
int n = option_find_int(options, "filters",1);
int size = option_find_int(options, "size",1);
int stride = option_find_int(options, "stride",1);
char *activation_s = option_find_str(options, "activation", "sigmoid");
ACTIVATION activation = get_activation(activation_s);
if(count == 0){
h = option_find_int(options, "height",1);
w = option_find_int(options, "width",1);
c = option_find_int(options, "channels",1);
}else{
image m = get_network_image_layer(net, count-1);
h = m.h;
w = m.w;
c = m.c;
if(h == 0) error("Layer before convolutional layer must output image.");
}
convolutional_layer *layer = make_convolutional_layer(h,w,c,n,size,stride, activation);
char *data = option_find_str(options, "data", 0);
if(data){
char *curr = data;
char *next = data;
for(i = 0; i < n; ++i){
while(*++next !='\0' && *next != ',');
*next = '\0';
sscanf(curr, "%g", &layer->biases[i]);
curr = next+1;
}
for(i = 0; i < c*n*size*size; ++i){
while(*++next !='\0' && *next != ',');
*next = '\0';
sscanf(curr, "%g", &layer->filters[i]);
curr = next+1;
}
}
option_unused(options);
return layer;
}
connected_layer *parse_connected(list *options, network net, int count)
{
int i;
int input;
int output = option_find_int(options, "output",1);
char *activation_s = option_find_str(options, "activation", "sigmoid");
ACTIVATION activation = get_activation(activation_s);
if(count == 0){
input = option_find_int(options, "input",1);
}else{
input = get_network_output_size_layer(net, count-1);
}
connected_layer *layer = make_connected_layer(input, output, activation);
char *data = option_find_str(options, "data", 0);
if(data){
char *curr = data;
char *next = data;
for(i = 0; i < output; ++i){
while(*++next !='\0' && *next != ',');
*next = '\0';
sscanf(curr, "%g", &layer->biases[i]);
curr = next+1;
}
for(i = 0; i < input*output; ++i){
while(*++next !='\0' && *next != ',');
*next = '\0';
sscanf(curr, "%g", &layer->weights[i]);
curr = next+1;
}
}
option_unused(options);
return layer;
}
softmax_layer *parse_softmax(list *options, network net, int count)
{
int input;
if(count == 0){
input = option_find_int(options, "input",1);
}else{
input = get_network_output_size_layer(net, count-1);
}
softmax_layer *layer = make_softmax_layer(input);
option_unused(options);
return layer;
}
maxpool_layer *parse_maxpool(list *options, network net, int count)
{
int h,w,c;
int stride = option_find_int(options, "stride",1);
if(count == 0){
h = option_find_int(options, "height",1);
w = option_find_int(options, "width",1);
c = option_find_int(options, "channels",1);
}else{
image m = get_network_image_layer(net, count-1);
h = m.h;
w = m.w;
c = m.c;
if(h == 0) error("Layer before convolutional layer must output image.");
}
maxpool_layer *layer = make_maxpool_layer(h,w,c,stride);
option_unused(options);
return layer;
}
network parse_network_cfg(char *filename)
{
@ -35,78 +159,29 @@ network parse_network_cfg(char *filename)
section *s = (section *)n->val;
list *options = s->options;
if(is_convolutional(s)){
int h,w,c;
int n = option_find_int(options, "filters",1);
int size = option_find_int(options, "size",1);
int stride = option_find_int(options, "stride",1);
char *activation_s = option_find_str(options, "activation", "sigmoid");
ACTIVATION activation = get_activation(activation_s);
if(count == 0){
h = option_find_int(options, "height",1);
w = option_find_int(options, "width",1);
c = option_find_int(options, "channels",1);
}else{
image m = get_network_image_layer(net, count-1);
h = m.h;
w = m.w;
c = m.c;
if(h == 0) error("Layer before convolutional layer must output image.");
}
convolutional_layer *layer = make_convolutional_layer(h,w,c,n,size,stride, activation);
convolutional_layer *layer = parse_convolutional(options, net, count);
net.types[count] = CONVOLUTIONAL;
net.layers[count] = layer;
option_unused(options);
}
else if(is_connected(s)){
int input;
int output = option_find_int(options, "output",1);
char *activation_s = option_find_str(options, "activation", "sigmoid");
ACTIVATION activation = get_activation(activation_s);
if(count == 0){
input = option_find_int(options, "input",1);
}else{
input = get_network_output_size_layer(net, count-1);
}
connected_layer *layer = make_connected_layer(input, output, activation);
}else if(is_connected(s)){
connected_layer *layer = parse_connected(options, net, count);
net.types[count] = CONNECTED;
net.layers[count] = layer;
option_unused(options);
}else if(is_softmax(s)){
int input;
if(count == 0){
input = option_find_int(options, "input",1);
}else{
input = get_network_output_size_layer(net, count-1);
}
softmax_layer *layer = make_softmax_layer(input);
softmax_layer *layer = parse_softmax(options, net, count);
net.types[count] = SOFTMAX;
net.layers[count] = layer;
option_unused(options);
}else if(is_maxpool(s)){
int h,w,c;
int stride = option_find_int(options, "stride",1);
//char *activation_s = option_find_str(options, "activation", "sigmoid");
if(count == 0){
h = option_find_int(options, "height",1);
w = option_find_int(options, "width",1);
c = option_find_int(options, "channels",1);
}else{
image m = get_network_image_layer(net, count-1);
h = m.h;
w = m.w;
c = m.c;
if(h == 0) error("Layer before convolutional layer must output image.");
}
maxpool_layer *layer = make_maxpool_layer(h,w,c,stride);
maxpool_layer *layer = parse_maxpool(options, net, count);
net.types[count] = MAXPOOL;
net.layers[count] = layer;
option_unused(options);
}else{
fprintf(stderr, "Type not recognized: %s\n", s->type);
}
free_section(s);
++count;
n = n->next;
}
free_list(sections);
net.outputs = get_network_output_size(net);
net.output = get_network_output(net);
return net;