From ced198e9390195875d743d77eadece99c7fd5b38 Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Mon, 19 Mar 2018 23:17:26 +0000
Subject: [PATCH] Fixed gpu_id for DLL/SO
---
src/region_layer.c | 192 +++++++++++++++++++++++++++++++++++-------------
1 files changed, 140 insertions(+), 52 deletions(-)
diff --git a/src/region_layer.c b/src/region_layer.c
index ac30e88..f179906 100644
--- a/src/region_layer.c
+++ b/src/region_layer.c
@@ -9,7 +9,9 @@
#include <string.h>
#include <stdlib.h>
-region_layer make_region_layer(int batch, int w, int h, int n, int classes, int coords)
+#define DOABS 1
+
+region_layer make_region_layer(int batch, int w, int h, int n, int classes, int coords, int max_boxes)
{
region_layer l = {0};
l.type = REGION;
@@ -25,7 +27,8 @@
l.bias_updates = calloc(n*2, sizeof(float));
l.outputs = h*w*n*(classes + coords + 1);
l.inputs = l.outputs;
- l.truths = 30*(5);
+ l.max_boxes = max_boxes;
+ l.truths = max_boxes*(5);
l.delta = calloc(batch*l.outputs, sizeof(float));
l.output = calloc(batch*l.outputs, sizeof(float));
int i;
@@ -42,13 +45,36 @@
l.delta_gpu = cuda_make_array(l.delta, batch*l.outputs);
#endif
- fprintf(stderr, "Region Layer\n");
+ fprintf(stderr, "detection\n");
srand(0);
return l;
}
-#define DOABS 1
+void resize_region_layer(layer *l, int w, int h)
+{
+ int old_w = l->w;
+ int old_h = l->h;
+ l->w = w;
+ l->h = h;
+
+ l->outputs = h*w*l->n*(l->classes + l->coords + 1);
+ l->inputs = l->outputs;
+
+ l->output = realloc(l->output, l->batch*l->outputs*sizeof(float));
+ l->delta = realloc(l->delta, l->batch*l->outputs*sizeof(float));
+
+#ifdef GPU
+ if (old_w < w || old_h < h) {
+ cuda_free(l->delta_gpu);
+ cuda_free(l->output_gpu);
+
+ l->delta_gpu = cuda_make_array(l->delta, l->batch*l->outputs);
+ l->output_gpu = cuda_make_array(l->output, l->batch*l->outputs);
+ }
+#endif
+}
+
box get_region_box(float *x, float *biases, int n, int index, int i, int j, int w, int h)
{
box b;
@@ -84,28 +110,48 @@
return iou;
}
-void delta_region_class(float *output, float *delta, int index, int class, int classes, tree *hier, float scale, float *avg_cat)
+void delta_region_class(float *output, float *delta, int index, int class_id, int classes, tree *hier, float scale, float *avg_cat, int focal_loss)
{
int i, n;
if(hier){
float pred = 1;
- while(class >= 0){
- pred *= output[index + class];
- int g = hier->group[class];
+ while(class_id >= 0){
+ pred *= output[index + class_id];
+ int g = hier->group[class_id];
int offset = hier->group_offset[g];
for(i = 0; i < hier->group_size[g]; ++i){
delta[index + offset + i] = scale * (0 - output[index + offset + i]);
}
- delta[index + class] = scale * (1 - output[index + class]);
+ delta[index + class_id] = scale * (1 - output[index + class_id]);
- class = hier->parent[class];
+ class_id = hier->parent[class_id];
}
*avg_cat += pred;
- } else {
- for(n = 0; n < classes; ++n){
- delta[index + n] = scale * (((n == class)?1 : 0) - output[index + n]);
- if(n == class) *avg_cat += output[index + n];
- }
+ } else {
+ // Focal loss
+ if (focal_loss) {
+ // Focal Loss for Dense Object Detection: http://blog.csdn.net/linmingan/article/details/77885832
+ float alpha = 0.5; // 0.25 or 0.5
+ //float gamma = 2; // hardcoded in many places of the grad-formula
+
+ int ti = index + class_id;
+ float grad = -2 * (1 - output[ti])*logf(fmaxf(output[ti], 0.0000001))*output[ti] + (1 - output[ti])*(1 - output[ti]);
+
+ for (n = 0; n < classes; ++n) {
+ delta[index + n] = scale * (((n == class_id) ? 1 : 0) - output[index + n]);
+
+ delta[index + n] *= alpha*grad;
+
+ if (n == class_id) *avg_cat += output[index + n];
+ }
+ }
+ else {
+ // default
+ for (n = 0; n < classes; ++n) {
+ delta[index + n] = scale * (((n == class_id) ? 1 : 0) - output[index + n]);
+ if (n == class_id) *avg_cat += output[index + n];
+ }
+ }
}
}
@@ -125,7 +171,9 @@
int i,j,b,t,n;
int size = l.coords + l.classes + 1;
memcpy(l.output, state.input, l.outputs*l.batch*sizeof(float));
- reorg(l.output, l.w*l.h, size*l.n, l.batch, 1);
+ #ifndef GPU
+ flatten(l.output, l.w*l.h, size*l.n, l.batch, 1);
+ #endif
for (b = 0; b < l.batch; ++b){
for(i = 0; i < l.h*l.w*l.n; ++i){
int index = size*i + b*l.outputs;
@@ -134,33 +182,23 @@
}
+#ifndef GPU
if (l.softmax_tree){
-#ifdef GPU
- cuda_push_array(l.output_gpu, l.output, l.batch*l.outputs);
- int i;
- int count = 5;
- for (i = 0; i < l.softmax_tree->groups; ++i) {
- int group_size = l.softmax_tree->group_size[i];
- softmax_gpu(l.output_gpu+count, group_size, l.classes + 5, l.w*l.h*l.n*l.batch, 1, l.output_gpu + count);
- count += group_size;
- }
- cuda_pull_array(l.output_gpu, l.output, l.batch*l.outputs);
-#else
for (b = 0; b < l.batch; ++b){
for(i = 0; i < l.h*l.w*l.n; ++i){
int index = size*i + b*l.outputs;
softmax_tree(l.output + index + 5, 1, 0, 1, l.softmax_tree, l.output + index + 5);
}
}
-#endif
} else if (l.softmax){
for (b = 0; b < l.batch; ++b){
for(i = 0; i < l.h*l.w*l.n; ++i){
int index = size*i + b*l.outputs;
- softmax(l.output + index + 5, l.classes, 1, l.output + index + 5);
+ softmax(l.output + index + 5, l.classes, 1, l.output + index + 5, 1);
}
}
}
+#endif
if(!state.train) return;
memset(l.delta, 0, l.outputs * l.batch * sizeof(float));
float avg_iou = 0;
@@ -172,19 +210,46 @@
int class_count = 0;
*(l.cost) = 0;
for (b = 0; b < l.batch; ++b) {
+ if(l.softmax_tree){
+ int onlyclass_id = 0;
+ for(t = 0; t < l.max_boxes; ++t){
+ box truth = float_to_box(state.truth + t*5 + b*l.truths);
+ if(!truth.x) break;
+ int class_id = state.truth[t*5 + b*l.truths + 4];
+ float maxp = 0;
+ int maxi = 0;
+ if(truth.x > 100000 && truth.y > 100000){
+ for(n = 0; n < l.n*l.w*l.h; ++n){
+ int index = size*n + b*l.outputs + 5;
+ float scale = l.output[index-1];
+ float p = scale*get_hierarchy_probability(l.output + index, l.softmax_tree, class_id);
+ if(p > maxp){
+ maxp = p;
+ maxi = n;
+ }
+ }
+ int index = size*maxi + b*l.outputs + 5;
+ delta_region_class(l.output, l.delta, index, class_id, l.classes, l.softmax_tree, l.class_scale, &avg_cat, l.focal_loss);
+ ++class_count;
+ onlyclass_id = 1;
+ break;
+ }
+ }
+ if(onlyclass_id) continue;
+ }
for (j = 0; j < l.h; ++j) {
for (i = 0; i < l.w; ++i) {
for (n = 0; n < l.n; ++n) {
int index = size*(j*l.w*l.n + i*l.n + n) + b*l.outputs;
box pred = get_region_box(l.output, l.biases, n, index, i, j, l.w, l.h);
float best_iou = 0;
- int best_class = -1;
- for(t = 0; t < 30; ++t){
+ int best_class_id = -1;
+ for(t = 0; t < l.max_boxes; ++t){
box truth = float_to_box(state.truth + t*5 + b*l.truths);
if(!truth.x) break;
float iou = box_iou(pred, truth);
if (iou > best_iou) {
- best_class = state.truth[t*5 + b*l.truths + 4];
+ best_class_id = state.truth[t*5 + b*l.truths + 4];
best_iou = iou;
}
}
@@ -195,7 +260,7 @@
if (best_iou > l.thresh) {
l.delta[index + 4] = 0;
if(l.classfix > 0){
- delta_region_class(l.output, l.delta, index + 5, best_class, l.classes, l.softmax_tree, l.class_scale*(l.classfix == 2 ? l.output[index + 4] : 1), &avg_cat);
+ delta_region_class(l.output, l.delta, index + 5, best_class_id, l.classes, l.softmax_tree, l.class_scale*(l.classfix == 2 ? l.output[index + 4] : 1), &avg_cat, l.focal_loss);
++class_count;
}
}
@@ -216,7 +281,7 @@
}
}
}
- for(t = 0; t < 30; ++t){
+ for(t = 0; t < l.max_boxes; ++t){
box truth = float_to_box(state.truth + t*5 + b*l.truths);
if(!truth.x) break;
@@ -265,15 +330,17 @@
}
- int class = state.truth[t*5 + b*l.truths + 4];
- if (l.map) class = l.map[class];
- delta_region_class(l.output, l.delta, best_index + 5, class, l.classes, l.softmax_tree, l.class_scale, &avg_cat);
+ int class_id = state.truth[t*5 + b*l.truths + 4];
+ if (l.map) class_id = l.map[class_id];
+ delta_region_class(l.output, l.delta, best_index + 5, class_id, l.classes, l.softmax_tree, l.class_scale, &avg_cat, l.focal_loss);
++count;
++class_count;
}
}
//printf("\n");
- reorg(l.delta, l.w*l.h, size*l.n, l.batch, 0);
+ #ifndef GPU
+ flatten(l.delta, l.w*l.h, size*l.n, l.batch, 0);
+ #endif
*(l.cost) = pow(mag_array(l.delta, l.outputs * l.batch), 2);
printf("Region Avg IOU: %f, Class: %f, Obj: %f, No Obj: %f, Avg Recall: %f, count: %d\n", avg_iou/count, avg_cat/class_count, avg_obj/count, avg_anyobj/(l.w*l.h*l.n*l.batch), recall/count, count);
}
@@ -283,7 +350,7 @@
axpy_cpu(l.batch*l.inputs, 1, l.delta, 1, state.delta, 1);
}
-void get_region_boxes(layer l, int w, int h, float thresh, float **probs, box *boxes, int only_objectness)
+void get_region_boxes(layer l, int w, int h, float thresh, float **probs, box *boxes, int only_objectness, int *map)
{
int i,j,n;
float *predictions = l.output;
@@ -307,16 +374,23 @@
hierarchy_predictions(predictions + class_index, l.classes, l.softmax_tree, 0);
int found = 0;
- for(j = l.classes - 1; j >= 0; --j){
- if(!found && predictions[class_index + j] > .5){
- found = 1;
- } else {
- predictions[class_index + j] = 0;
+ if(map){
+ for(j = 0; j < 200; ++j){
+ float prob = scale*predictions[class_index+map[j]];
+ probs[index][j] = (prob > thresh) ? prob : 0;
}
- float prob = predictions[class_index+j];
- probs[index][j] = (scale > thresh) ? prob : 0;
+ } else {
+ for(j = l.classes - 1; j >= 0; --j){
+ if(!found && predictions[class_index + j] > .5){
+ found = 1;
+ } else {
+ predictions[class_index + j] = 0;
+ }
+ float prob = predictions[class_index+j];
+ probs[index][j] = (scale > thresh) ? prob : 0;
+ }
}
- }else{
+ } else {
for(j = 0; j < l.classes; ++j){
float prob = scale*predictions[class_index+j];
probs[index][j] = (prob > thresh) ? prob : 0;
@@ -339,6 +413,18 @@
return;
}
*/
+ flatten_ongpu(state.input, l.h*l.w, l.n*(l.coords + l.classes + 1), l.batch, 1, l.output_gpu);
+ if(l.softmax_tree){
+ int i;
+ int count = 5;
+ for (i = 0; i < l.softmax_tree->groups; ++i) {
+ int group_size = l.softmax_tree->group_size[i];
+ softmax_gpu(l.output_gpu+count, group_size, l.classes + 5, l.w*l.h*l.n*l.batch, 1, l.output_gpu + count);
+ count += group_size;
+ }
+ }else if (l.softmax){
+ softmax_gpu(l.output_gpu+5, l.classes, l.classes + 5, l.w*l.h*l.n*l.batch, 1, l.output_gpu + 5);
+ }
float *in_cpu = calloc(l.batch*l.inputs, sizeof(float));
float *truth_cpu = 0;
@@ -347,22 +433,24 @@
truth_cpu = calloc(num_truth, sizeof(float));
cuda_pull_array(state.truth, truth_cpu, num_truth);
}
- cuda_pull_array(state.input, in_cpu, l.batch*l.inputs);
+ cuda_pull_array(l.output_gpu, in_cpu, l.batch*l.inputs);
+ //cudaStreamSynchronize(get_cuda_stream());
network_state cpu_state = state;
cpu_state.train = state.train;
cpu_state.truth = truth_cpu;
cpu_state.input = in_cpu;
forward_region_layer(l, cpu_state);
- cuda_push_array(l.output_gpu, l.output, l.batch*l.outputs);
- cuda_push_array(l.delta_gpu, l.delta, l.batch*l.outputs);
+ //cuda_push_array(l.output_gpu, l.output, l.batch*l.outputs);
free(cpu_state.input);
+ if(!state.train) return;
+ cuda_push_array(l.delta_gpu, l.delta, l.batch*l.outputs);
+ //cudaStreamSynchronize(get_cuda_stream());
if(cpu_state.truth) free(cpu_state.truth);
}
void backward_region_layer_gpu(region_layer l, network_state state)
{
- axpy_ongpu(l.batch*l.outputs, 1, l.delta_gpu, 1, state.delta, 1);
- //copy_ongpu(l.batch*l.inputs, l.delta_gpu, 1, state.delta, 1);
+ flatten_ongpu(l.delta_gpu, l.h*l.w, l.n*(l.coords + l.classes + 1), l.batch, 0, state.delta);
}
#endif
--
Gitblit v1.10.0