From 132251d72325e1005ef6c47f83d6a4e9b9355d12 Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Sun, 03 Jun 2018 21:37:08 +0000
Subject: [PATCH] You can do: fuse_conv_batchnorm(network net) from DLL/SO-library
---
src/parser.c | 107 +++++++++++++++++++++++++++++++++++++++++++++++++++--
1 files changed, 102 insertions(+), 5 deletions(-)
diff --git a/src/parser.c b/src/parser.c
index 1adcd16..c1ee98c 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -30,6 +30,8 @@
#include "shortcut_layer.h"
#include "softmax_layer.h"
#include "utils.h"
+#include "upsample_layer.h"
+#include "yolo_layer.h"
#include <stdint.h>
typedef struct{
@@ -47,6 +49,7 @@
if (strcmp(type, "[cost]")==0) return COST;
if (strcmp(type, "[detection]")==0) return DETECTION;
if (strcmp(type, "[region]")==0) return REGION;
+ if (strcmp(type, "[yolo]") == 0) return YOLO;
if (strcmp(type, "[local]")==0) return LOCAL;
if (strcmp(type, "[conv]")==0
|| strcmp(type, "[convolutional]")==0) return CONVOLUTIONAL;
@@ -71,6 +74,7 @@
if (strcmp(type, "[soft]")==0
|| strcmp(type, "[softmax]")==0) return SOFTMAX;
if (strcmp(type, "[route]")==0) return ROUTE;
+ if (strcmp(type, "[upsample]") == 0) return UPSAMPLE;
return BLANK;
}
@@ -235,21 +239,92 @@
return layer;
}
+int *parse_yolo_mask(char *a, int *num)
+{
+ int *mask = 0;
+ if (a) {
+ int len = strlen(a);
+ int n = 1;
+ int i;
+ for (i = 0; i < len; ++i) {
+ if (a[i] == ',') ++n;
+ }
+ mask = calloc(n, sizeof(int));
+ for (i = 0; i < n; ++i) {
+ int val = atoi(a);
+ mask[i] = val;
+ a = strchr(a, ',') + 1;
+ }
+ *num = n;
+ }
+ return mask;
+}
+
+layer parse_yolo(list *options, size_params params)
+{
+ int classes = option_find_int(options, "classes", 20);
+ int total = option_find_int(options, "num", 1);
+ int num = total;
+
+ char *a = option_find_str(options, "mask", 0);
+ int *mask = parse_yolo_mask(a, &num);
+ int max_boxes = option_find_int_quiet(options, "max", 90);
+ layer l = make_yolo_layer(params.batch, params.w, params.h, num, total, mask, classes, max_boxes);
+ if (l.outputs != params.inputs) {
+ printf("Error: l.outputs == params.inputs \n");
+ printf("filters= in the [convolutional]-layer doesn't correspond to classes= or mask= in [yolo]-layer \n");
+ exit(EXIT_FAILURE);
+ }
+ //assert(l.outputs == params.inputs);
+
+ //l.max_boxes = option_find_int_quiet(options, "max", 90);
+ l.jitter = option_find_float(options, "jitter", .2);
+ l.focal_loss = option_find_int_quiet(options, "focal_loss", 0);
+
+ l.ignore_thresh = option_find_float(options, "ignore_thresh", .5);
+ l.truth_thresh = option_find_float(options, "truth_thresh", 1);
+ l.random = option_find_int_quiet(options, "random", 0);
+
+ char *map_file = option_find_str(options, "map", 0);
+ if (map_file) l.map = read_map(map_file);
+
+ a = option_find_str(options, "anchors", 0);
+ if (a) {
+ int len = strlen(a);
+ int n = 1;
+ int i;
+ for (i = 0; i < len; ++i) {
+ if (a[i] == ',') ++n;
+ }
+ for (i = 0; i < n && i < total*2; ++i) {
+ float bias = atof(a);
+ l.biases[i] = bias;
+ a = strchr(a, ',') + 1;
+ }
+ }
+ return l;
+}
+
layer parse_region(list *options, size_params params)
{
int coords = option_find_int(options, "coords", 4);
int classes = option_find_int(options, "classes", 20);
int num = option_find_int(options, "num", 1);
- int max_boxes = option_find_int_quiet(options, "max", 30);
+ int max_boxes = option_find_int_quiet(options, "max", 90);
layer l = make_region_layer(params.batch, params.w, params.h, num, classes, coords, max_boxes);
- assert(l.outputs == params.inputs);
+ if (l.outputs != params.inputs) {
+ printf("Error: l.outputs == params.inputs \n");
+ printf("filters= in the [convolutional]-layer doesn't correspond to classes= or num= in [region]-layer \n");
+ exit(EXIT_FAILURE);
+ }
+ //assert(l.outputs == params.inputs);
l.log = option_find_int_quiet(options, "log", 0);
l.sqrt = option_find_int_quiet(options, "sqrt", 0);
- l.small_object = option_find_int_quiet(options, "small_object", 0);
l.softmax = option_find_int(options, "softmax", 0);
+ l.focal_loss = option_find_int_quiet(options, "focal_loss", 0);
//l.max_boxes = option_find_int_quiet(options, "max",30);
l.jitter = option_find_float(options, "jitter", .2);
l.rescore = option_find_int_quiet(options, "rescore",0);
@@ -262,6 +337,7 @@
l.coord_scale = option_find_float(options, "coord_scale", 1);
l.object_scale = option_find_float(options, "object_scale", 1);
l.noobject_scale = option_find_float(options, "noobject_scale", 1);
+ l.mask_scale = option_find_float(options, "mask_scale", 1);
l.class_scale = option_find_float(options, "class_scale", 1);
l.bias_match = option_find_int_quiet(options, "bias_match",0);
@@ -278,7 +354,7 @@
for(i = 0; i < len; ++i){
if (a[i] == ',') ++n;
}
- for(i = 0; i < n; ++i){
+ for(i = 0; i < n && i < num*2; ++i){
float bias = atof(a);
l.biases[i] = bias;
a = strchr(a, ',')+1;
@@ -468,6 +544,15 @@
return l;
}
+layer parse_upsample(list *options, size_params params, network net)
+{
+
+ int stride = option_find_int(options, "stride", 2);
+ layer l = make_upsample_layer(params.batch, params.w, params.h, params.c, stride);
+ l.scale = option_find_float_quiet(options, "scale", 1);
+ return l;
+}
+
route_layer parse_route(list *options, size_params params, network net)
{
char *l = option_find(options, "layers");
@@ -547,7 +632,9 @@
net->inputs = option_find_int_quiet(options, "inputs", net->h * net->w * net->c);
net->max_crop = option_find_int_quiet(options, "max_crop",net->w*2);
net->min_crop = option_find_int_quiet(options, "min_crop",net->w);
+ net->flip = option_find_int_quiet(options, "flip", 1);
+ net->small_object = option_find_int_quiet(options, "small_object", 0);
net->angle = option_find_float_quiet(options, "angle", 0);
net->aspect = option_find_float_quiet(options, "aspect", 1);
net->saturation = option_find_float_quiet(options, "saturation", 1);
@@ -560,6 +647,9 @@
char *policy_s = option_find_str(options, "policy", "constant");
net->policy = get_policy(policy_s);
net->burn_in = option_find_int_quiet(options, "burn_in", 0);
+#ifdef CUDNN_HALF
+ net->burn_in = 0;
+#endif
if(net->policy == STEP){
net->step = option_find_int(options, "step", 1);
net->scale = option_find_float(options, "scale", 1);
@@ -632,6 +722,7 @@
params.time_steps = net.time_steps;
params.net = net;
+ float bflops = 0;
size_t workspace_size = 0;
n = n->next;
int count = 0;
@@ -639,7 +730,7 @@
fprintf(stderr, "layer filters size input output\n");
while(n){
params.index = count;
- fprintf(stderr, "%5d ", count);
+ fprintf(stderr, "%4d ", count);
s = (section *)n->val;
options = s->options;
layer l = {0};
@@ -664,6 +755,8 @@
l = parse_cost(options, params);
}else if(lt == REGION){
l = parse_region(options, params);
+ }else if (lt == YOLO) {
+ l = parse_yolo(options, params);
}else if(lt == DETECTION){
l = parse_detection(options, params);
}else if(lt == SOFTMAX){
@@ -683,6 +776,8 @@
l = parse_avgpool(options, params);
}else if(lt == ROUTE){
l = parse_route(options, params, net);
+ }else if (lt == UPSAMPLE) {
+ l = parse_upsample(options, params, net);
}else if(lt == SHORTCUT){
l = parse_shortcut(options, params, net);
}else if(lt == DROPOUT){
@@ -712,10 +807,12 @@
params.c = l.out_c;
params.inputs = l.outputs;
}
+ if (l.bflops > 0) bflops += l.bflops;
}
free_list(sections);
net.outputs = get_network_output_size(net);
net.output = get_network_output(net);
+ printf("Total BFLOPS %5.3f \n", bflops);
if(workspace_size){
//printf("%ld\n", workspace_size);
#ifdef GPU
--
Gitblit v1.10.0