2013-11-04 23:11:01 +04:00
|
|
|
#include "connected_layer.h"
|
|
|
|
#include "convolutional_layer.h"
|
|
|
|
#include "maxpool_layer.h"
|
|
|
|
#include "network.h"
|
|
|
|
#include "image.h"
|
2013-11-07 04:09:41 +04:00
|
|
|
#include "parser.h"
|
2013-11-13 22:50:38 +04:00
|
|
|
#include "data.h"
|
|
|
|
#include "matrix.h"
|
2013-12-03 04:41:40 +04:00
|
|
|
#include "utils.h"
|
2014-01-25 02:49:02 +04:00
|
|
|
#include "mini_blas.h"
|
2013-11-04 23:11:01 +04:00
|
|
|
|
|
|
|
#include <time.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <stdio.h>
|
|
|
|
|
2014-01-29 04:28:42 +04:00
|
|
|
#define _GNU_SOURCE
|
|
|
|
#include <fenv.h>
|
|
|
|
|
2013-11-04 23:11:01 +04:00
|
|
|
void test_convolve()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("dog.jpg",300,400);
|
|
|
|
printf("dog channels %d\n", dog.c);
|
|
|
|
image kernel = make_random_image(3,3,dog.c);
|
|
|
|
image edge = make_image(dog.h, dog.w, 1);
|
|
|
|
int i;
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
for(i = 0; i < 1000; ++i){
|
|
|
|
convolve(dog, kernel, 1, 0, edge, 1);
|
|
|
|
}
|
|
|
|
end = clock();
|
|
|
|
printf("Convolutions: %lf seconds\n", (float)(end-start)/CLOCKS_PER_SEC);
|
|
|
|
show_image_layers(edge, "Test Convolve");
|
2013-11-04 23:11:01 +04:00
|
|
|
}
|
|
|
|
|
2014-08-28 06:11:46 +04:00
|
|
|
#ifdef GPU
|
|
|
|
|
|
|
|
void test_convolutional_layer()
|
|
|
|
{
|
|
|
|
int i;
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("data/dog.jpg",224,224);
|
|
|
|
network net = parse_network_cfg("cfg/convolutional.cfg");
|
|
|
|
// data test = load_cifar10_data("data/cifar10/test_batch.bin");
|
|
|
|
// float *X = calloc(net.batch*test.X.cols, sizeof(float));
|
|
|
|
// float *y = calloc(net.batch*test.y.cols, sizeof(float));
|
2014-08-28 06:11:46 +04:00
|
|
|
int in_size = get_network_input_size(net)*net.batch;
|
2014-10-13 11:29:01 +04:00
|
|
|
int del_size = get_network_output_size_layer(net, 0)*net.batch;
|
2014-08-28 06:11:46 +04:00
|
|
|
int size = get_network_output_size(net)*net.batch;
|
2014-10-13 11:29:01 +04:00
|
|
|
float *X = calloc(in_size, sizeof(float));
|
|
|
|
float *y = calloc(size, sizeof(float));
|
2014-08-28 06:11:46 +04:00
|
|
|
for(i = 0; i < in_size; ++i){
|
|
|
|
X[i] = dog.data[i%get_network_input_size(net)];
|
|
|
|
}
|
2014-11-22 02:35:19 +03:00
|
|
|
// get_batch(test, net.batch, X, y);
|
2014-08-28 06:11:46 +04:00
|
|
|
clock_t start, end;
|
|
|
|
cl_mem input_cl = cl_make_array(X, in_size);
|
2014-10-13 11:29:01 +04:00
|
|
|
cl_mem truth_cl = cl_make_array(y, size);
|
2014-08-28 06:11:46 +04:00
|
|
|
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network_gpu(net, input_cl, truth_cl, 1);
|
2014-08-28 06:11:46 +04:00
|
|
|
start = clock();
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network_gpu(net, input_cl, truth_cl, 1);
|
2014-08-28 06:11:46 +04:00
|
|
|
end = clock();
|
|
|
|
float gpu_sec = (float)(end-start)/CLOCKS_PER_SEC;
|
2014-10-13 11:29:01 +04:00
|
|
|
printf("forward gpu: %f sec\n", gpu_sec);
|
|
|
|
start = clock();
|
|
|
|
backward_network_gpu(net, input_cl);
|
|
|
|
end = clock();
|
|
|
|
gpu_sec = (float)(end-start)/CLOCKS_PER_SEC;
|
|
|
|
printf("backward gpu: %f sec\n", gpu_sec);
|
|
|
|
//float gpu_cost = get_network_cost(net);
|
2014-08-28 06:11:46 +04:00
|
|
|
float *gpu_out = calloc(size, sizeof(float));
|
|
|
|
memcpy(gpu_out, get_network_output(net), size*sizeof(float));
|
|
|
|
|
2014-10-13 11:29:01 +04:00
|
|
|
float *gpu_del = calloc(del_size, sizeof(float));
|
|
|
|
memcpy(gpu_del, get_network_delta_layer(net, 0), del_size*sizeof(float));
|
|
|
|
|
2014-11-22 02:35:19 +03:00
|
|
|
/*
|
|
|
|
start = clock();
|
|
|
|
forward_network(net, X, y, 1);
|
|
|
|
backward_network(net, X);
|
|
|
|
float cpu_cost = get_network_cost(net);
|
|
|
|
end = clock();
|
|
|
|
float cpu_sec = (float)(end-start)/CLOCKS_PER_SEC;
|
|
|
|
float *cpu_out = calloc(size, sizeof(float));
|
|
|
|
memcpy(cpu_out, get_network_output(net), size*sizeof(float));
|
|
|
|
float *cpu_del = calloc(del_size, sizeof(float));
|
|
|
|
memcpy(cpu_del, get_network_delta_layer(net, 0), del_size*sizeof(float));
|
|
|
|
|
|
|
|
float sum = 0;
|
|
|
|
float del_sum = 0;
|
|
|
|
for(i = 0; i < size; ++i) sum += pow(gpu_out[i] - cpu_out[i], 2);
|
|
|
|
for(i = 0; i < del_size; ++i) {
|
|
|
|
//printf("%f %f\n", cpu_del[i], gpu_del[i]);
|
|
|
|
del_sum += pow(cpu_del[i] - gpu_del[i], 2);
|
2014-08-28 06:11:46 +04:00
|
|
|
}
|
2014-10-13 11:29:01 +04:00
|
|
|
printf("GPU cost: %f, CPU cost: %f\n", gpu_cost, cpu_cost);
|
|
|
|
printf("gpu: %f sec, cpu: %f sec, diff: %f, delta diff: %f, size: %d\n", gpu_sec, cpu_sec, sum, del_sum, size);
|
2014-11-22 02:35:19 +03:00
|
|
|
*/
|
2014-10-13 11:29:01 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
void test_col2im()
|
|
|
|
{
|
|
|
|
float col[] = {1,2,1,2,
|
2014-11-22 02:35:19 +03:00
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2,
|
|
|
|
1,2,1,2};
|
2014-10-13 11:29:01 +04:00
|
|
|
float im[16] = {0};
|
|
|
|
int batch = 1;
|
|
|
|
int channels = 1;
|
|
|
|
int height=4;
|
|
|
|
int width=4;
|
|
|
|
int ksize = 3;
|
|
|
|
int stride = 1;
|
|
|
|
int pad = 0;
|
|
|
|
col2im_gpu(col, batch,
|
2014-11-22 02:35:19 +03:00
|
|
|
channels, height, width,
|
|
|
|
ksize, stride, pad, im);
|
2014-10-13 11:29:01 +04:00
|
|
|
int i;
|
|
|
|
for(i = 0; i < 16; ++i)printf("%f,", im[i]);
|
|
|
|
printf("\n");
|
|
|
|
/*
|
2014-11-22 02:35:19 +03:00
|
|
|
float data_im[] = {
|
|
|
|
1,2,3,4,
|
|
|
|
5,6,7,8,
|
|
|
|
9,10,11,12
|
|
|
|
};
|
|
|
|
float data_col[18] = {0};
|
|
|
|
im2col_cpu(data_im, batch,
|
|
|
|
channels, height, width,
|
|
|
|
ksize, stride, pad, data_col) ;
|
|
|
|
for(i = 0; i < 18; ++i)printf("%f,", data_col[i]);
|
|
|
|
printf("\n");
|
|
|
|
*/
|
2014-08-28 06:11:46 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
2014-01-25 02:49:02 +04:00
|
|
|
void test_convolve_matrix()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("dog.jpg",300,400);
|
|
|
|
printf("dog channels %d\n", dog.c);
|
|
|
|
|
|
|
|
int size = 11;
|
|
|
|
int stride = 4;
|
|
|
|
int n = 40;
|
|
|
|
float *filters = make_random_image(size, size, dog.c*n).data;
|
|
|
|
|
|
|
|
int mw = ((dog.h-size)/stride+1)*((dog.w-size)/stride+1);
|
|
|
|
int mh = (size*size*dog.c);
|
|
|
|
float *matrix = calloc(mh*mw, sizeof(float));
|
|
|
|
|
|
|
|
image edge = make_image((dog.h-size)/stride+1, (dog.w-size)/stride+1, n);
|
|
|
|
|
|
|
|
int i;
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
for(i = 0; i < 1000; ++i){
|
|
|
|
im2col_cpu(dog.data,1, dog.c, dog.h, dog.w, size, stride, 0, matrix);
|
|
|
|
gemm(0,0,n,mw,mh,1,filters,mh,matrix,mw,1,edge.data,mw);
|
|
|
|
}
|
|
|
|
end = clock();
|
|
|
|
printf("Convolutions: %lf seconds\n", (float)(end-start)/CLOCKS_PER_SEC);
|
|
|
|
show_image_layers(edge, "Test Convolve");
|
|
|
|
cvWaitKey(0);
|
2014-01-25 02:49:02 +04:00
|
|
|
}
|
|
|
|
|
2013-11-04 23:11:01 +04:00
|
|
|
void test_color()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("test_color.png", 300, 400);
|
|
|
|
show_image_layers(dog, "Test Color");
|
2013-11-04 23:11:01 +04:00
|
|
|
}
|
|
|
|
|
2013-12-03 04:41:40 +04:00
|
|
|
void verify_convolutional_layer()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
srand(0);
|
|
|
|
int i;
|
|
|
|
int n = 1;
|
|
|
|
int stride = 1;
|
|
|
|
int size = 3;
|
|
|
|
float eps = .00000001;
|
|
|
|
image test = make_random_image(5,5, 1);
|
|
|
|
convolutional_layer layer = *make_convolutional_layer(1,test.h,test.w,test.c, n, size, stride, 0, RELU,0,0,0);
|
|
|
|
image out = get_convolutional_image(layer);
|
|
|
|
float **jacobian = calloc(test.h*test.w*test.c, sizeof(float));
|
|
|
|
|
|
|
|
forward_convolutional_layer(layer, test.data);
|
|
|
|
image base = copy_image(out);
|
|
|
|
|
|
|
|
for(i = 0; i < test.h*test.w*test.c; ++i){
|
|
|
|
test.data[i] += eps;
|
|
|
|
forward_convolutional_layer(layer, test.data);
|
|
|
|
image partial = copy_image(out);
|
|
|
|
subtract_image(partial, base);
|
|
|
|
scale_image(partial, 1/eps);
|
|
|
|
jacobian[i] = partial.data;
|
|
|
|
test.data[i] -= eps;
|
|
|
|
}
|
|
|
|
float **jacobian2 = calloc(out.h*out.w*out.c, sizeof(float));
|
|
|
|
image in_delta = make_image(test.h, test.w, test.c);
|
|
|
|
image out_delta = get_convolutional_delta(layer);
|
|
|
|
for(i = 0; i < out.h*out.w*out.c; ++i){
|
|
|
|
out_delta.data[i] = 1;
|
|
|
|
backward_convolutional_layer(layer, in_delta.data);
|
|
|
|
image partial = copy_image(in_delta);
|
|
|
|
jacobian2[i] = partial.data;
|
|
|
|
out_delta.data[i] = 0;
|
|
|
|
}
|
|
|
|
int j;
|
|
|
|
float *j1 = calloc(test.h*test.w*test.c*out.h*out.w*out.c, sizeof(float));
|
|
|
|
float *j2 = calloc(test.h*test.w*test.c*out.h*out.w*out.c, sizeof(float));
|
|
|
|
for(i = 0; i < test.h*test.w*test.c; ++i){
|
|
|
|
for(j =0 ; j < out.h*out.w*out.c; ++j){
|
|
|
|
j1[i*out.h*out.w*out.c + j] = jacobian[i][j];
|
|
|
|
j2[i*out.h*out.w*out.c + j] = jacobian2[j][i];
|
|
|
|
printf("%f %f\n", jacobian[i][j], jacobian2[j][i]);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
image mj1 = float_to_image(test.w*test.h*test.c, out.w*out.h*out.c, 1, j1);
|
|
|
|
image mj2 = float_to_image(test.w*test.h*test.c, out.w*out.h*out.c, 1, j2);
|
|
|
|
printf("%f %f\n", avg_image_layer(mj1,0), avg_image_layer(mj2,0));
|
|
|
|
show_image(mj1, "forward jacobian");
|
|
|
|
show_image(mj2, "backward jacobian");
|
2013-12-03 04:41:40 +04:00
|
|
|
}
|
|
|
|
|
2013-11-04 23:11:01 +04:00
|
|
|
void test_load()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("dog.jpg", 300, 400);
|
|
|
|
show_image(dog, "Test Load");
|
|
|
|
show_image_layers(dog, "Test Load");
|
2013-11-04 23:11:01 +04:00
|
|
|
}
|
|
|
|
void test_upsample()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
image dog = load_image("dog.jpg", 300, 400);
|
|
|
|
int n = 3;
|
|
|
|
image up = make_image(n*dog.h, n*dog.w, dog.c);
|
|
|
|
upsample_image(dog, n, up);
|
|
|
|
show_image(up, "Test Upsample");
|
|
|
|
show_image_layers(up, "Test Upsample");
|
2013-11-04 23:11:01 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
void test_rotate()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
int i;
|
|
|
|
image dog = load_image("dog.jpg",300,400);
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
for(i = 0; i < 1001; ++i){
|
|
|
|
rotate_image(dog);
|
|
|
|
}
|
|
|
|
end = clock();
|
|
|
|
printf("Rotations: %lf seconds\n", (float)(end-start)/CLOCKS_PER_SEC);
|
|
|
|
show_image(dog, "Test Rotate");
|
|
|
|
|
|
|
|
image random = make_random_image(3,3,3);
|
|
|
|
show_image(random, "Test Rotate Random");
|
|
|
|
rotate_image(random);
|
|
|
|
show_image(random, "Test Rotate Random");
|
|
|
|
rotate_image(random);
|
|
|
|
show_image(random, "Test Rotate Random");
|
2013-11-04 23:11:01 +04:00
|
|
|
}
|
|
|
|
|
2013-11-13 22:50:38 +04:00
|
|
|
void test_parser()
|
2013-11-06 22:37:37 +04:00
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
network net = parse_network_cfg("cfg/trained_imagenet.cfg");
|
2014-11-19 00:51:04 +03:00
|
|
|
save_network(net, "cfg/trained_imagenet_smaller.cfg");
|
2013-11-13 22:50:38 +04:00
|
|
|
}
|
2013-11-06 22:37:37 +04:00
|
|
|
|
2014-11-06 01:49:58 +03:00
|
|
|
void train_asirra()
|
2013-11-07 04:09:41 +04:00
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
network net = parse_network_cfg("cfg/imagenet.cfg");
|
2014-10-22 01:49:18 +04:00
|
|
|
int imgs = 1000/net.batch+1;
|
|
|
|
//imgs = 1;
|
2014-11-22 02:35:19 +03:00
|
|
|
srand(2222222);
|
|
|
|
int i = 0;
|
|
|
|
char *labels[] = {"cat","dog"};
|
|
|
|
|
|
|
|
list *plist = get_paths("data/assira/train.list");
|
|
|
|
char **paths = (char **)list_to_array(plist);
|
|
|
|
int m = plist->size;
|
|
|
|
free_list(plist);
|
|
|
|
|
2014-10-25 22:57:26 +04:00
|
|
|
clock_t time;
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
while(1){
|
|
|
|
i += 1;
|
2014-10-25 22:57:26 +04:00
|
|
|
time=clock();
|
2014-11-22 02:35:19 +03:00
|
|
|
data train = load_data_random(imgs*net.batch, paths, m, labels, 2, 256, 256);
|
|
|
|
normalize_data_rows(train);
|
2014-10-25 22:57:26 +04:00
|
|
|
printf("Loaded: %lf seconds\n", sec(clock()-time));
|
|
|
|
time=clock();
|
2014-11-22 02:35:19 +03:00
|
|
|
//float loss = train_network_data(net, train, imgs);
|
2014-11-19 00:51:04 +03:00
|
|
|
float loss = 0;
|
2014-11-22 02:35:19 +03:00
|
|
|
printf("%d: %f, Time: %lf seconds\n", i*net.batch*imgs, loss, sec(clock()-time));
|
|
|
|
free_data(train);
|
|
|
|
if(i%10==0){
|
|
|
|
char buff[256];
|
|
|
|
sprintf(buff, "cfg/asirra_backup_%d.cfg", i);
|
|
|
|
save_network(net, buff);
|
|
|
|
}
|
|
|
|
//lr *= .99;
|
|
|
|
}
|
2013-11-13 22:50:38 +04:00
|
|
|
}
|
2014-04-11 12:00:27 +04:00
|
|
|
|
2014-11-22 02:35:19 +03:00
|
|
|
void train_detection_net()
|
|
|
|
{
|
|
|
|
float avg_loss = 1;
|
|
|
|
//network net = parse_network_cfg("/home/pjreddie/imagenet_backup/alexnet_1270.cfg");
|
|
|
|
network net = parse_network_cfg("cfg/detnet.cfg");
|
|
|
|
printf("Learning Rate: %g, Momentum: %g, Decay: %g\n", net.learning_rate, net.momentum, net.decay);
|
|
|
|
int imgs = 1000/net.batch+1;
|
2014-12-03 19:48:07 +03:00
|
|
|
//srand(time(0));
|
|
|
|
srand(23410);
|
2014-11-22 02:35:19 +03:00
|
|
|
int i = 0;
|
2014-11-28 21:38:26 +03:00
|
|
|
list *plist = get_paths("/home/pjreddie/data/imagenet/horse.txt");
|
2014-11-22 02:35:19 +03:00
|
|
|
char **paths = (char **)list_to_array(plist);
|
|
|
|
printf("%d\n", plist->size);
|
|
|
|
clock_t time;
|
|
|
|
while(1){
|
|
|
|
i += 1;
|
|
|
|
time=clock();
|
2014-11-28 21:38:26 +03:00
|
|
|
data train = load_data_detection_random(imgs*net.batch, paths, plist->size, 256, 256, 8, 8, 256);
|
2014-11-22 02:35:19 +03:00
|
|
|
//translate_data_rows(train, -144);
|
2014-12-03 19:48:07 +03:00
|
|
|
/*
|
|
|
|
image im = float_to_image(256, 256, 3, train.X.vals[0]);
|
|
|
|
float *truth = train.y.vals[0];
|
|
|
|
int j;
|
|
|
|
int r, c;
|
|
|
|
for(r = 0; r < 8; ++r){
|
|
|
|
for(c = 0; c < 8; ++c){
|
|
|
|
j = (r*8 + c) * 5;
|
|
|
|
if(truth[j]){
|
|
|
|
int d = 256/8;
|
|
|
|
int y = r*d+truth[j+1]*d;
|
|
|
|
int x = c*d+truth[j+2]*d;
|
|
|
|
int h = truth[j+3]*256;
|
|
|
|
int w = truth[j+4]*256;
|
|
|
|
printf("%f %f %f %f\n", truth[j+1], truth[j+2], truth[j+3], truth[j+4]);
|
|
|
|
printf("%d %d %d %d\n", x, y, w, h);
|
|
|
|
printf("%d %d %d %d\n", x-w/2, y-h/2, x+w/2, y+h/2);
|
|
|
|
draw_box(im, x-w/2, y-h/2, x+w/2, y+h/2);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
show_image(im, "box");
|
|
|
|
cvWaitKey(0);
|
|
|
|
*/
|
|
|
|
|
2014-11-22 02:35:19 +03:00
|
|
|
normalize_data_rows(train);
|
|
|
|
printf("Loaded: %lf seconds\n", sec(clock()-time));
|
|
|
|
time=clock();
|
|
|
|
#ifdef GPU
|
|
|
|
float loss = train_network_data_gpu(net, train, imgs);
|
|
|
|
avg_loss = avg_loss*.9 + loss*.1;
|
|
|
|
printf("%d: %f, %f avg, %lf seconds, %d images\n", i, loss, avg_loss, sec(clock()-time), i*imgs*net.batch);
|
|
|
|
#endif
|
|
|
|
free_data(train);
|
|
|
|
if(i%10==0){
|
|
|
|
char buff[256];
|
2014-12-03 19:48:07 +03:00
|
|
|
sprintf(buff, "/home/pjreddie/imagenet_backup/detnet_%d.cfg", i);
|
2014-11-22 02:35:19 +03:00
|
|
|
save_network(net, buff);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2014-10-25 22:57:26 +04:00
|
|
|
void train_imagenet()
|
|
|
|
{
|
2014-11-06 01:49:58 +03:00
|
|
|
float avg_loss = 1;
|
2014-11-22 02:35:19 +03:00
|
|
|
//network net = parse_network_cfg("/home/pjreddie/imagenet_backup/alexnet_1270.cfg");
|
2014-12-03 19:48:07 +03:00
|
|
|
network net = parse_network_cfg("cfg/alexnet.part");
|
2014-10-25 22:57:26 +04:00
|
|
|
printf("Learning Rate: %g, Momentum: %g, Decay: %g\n", net.learning_rate, net.momentum, net.decay);
|
|
|
|
int imgs = 1000/net.batch+1;
|
2014-11-22 02:35:19 +03:00
|
|
|
srand(time(0));
|
|
|
|
int i = 0;
|
2014-10-25 22:57:26 +04:00
|
|
|
char **labels = get_labels("/home/pjreddie/data/imagenet/cls.labels.list");
|
2014-10-28 05:45:06 +03:00
|
|
|
list *plist = get_paths("/data/imagenet/cls.train.list");
|
2014-10-25 22:57:26 +04:00
|
|
|
char **paths = (char **)list_to_array(plist);
|
2014-10-28 05:45:06 +03:00
|
|
|
printf("%d\n", plist->size);
|
2014-10-25 22:57:26 +04:00
|
|
|
clock_t time;
|
2014-11-22 02:35:19 +03:00
|
|
|
while(1){
|
|
|
|
i += 1;
|
2014-10-25 22:57:26 +04:00
|
|
|
time=clock();
|
2014-11-22 02:35:19 +03:00
|
|
|
data train = load_data_random(imgs*net.batch, paths, plist->size, labels, 1000, 256, 256);
|
2014-11-06 01:49:58 +03:00
|
|
|
//translate_data_rows(train, -144);
|
|
|
|
normalize_data_rows(train);
|
2014-10-25 22:57:26 +04:00
|
|
|
printf("Loaded: %lf seconds\n", sec(clock()-time));
|
|
|
|
time=clock();
|
2014-11-22 02:35:19 +03:00
|
|
|
#ifdef GPU
|
|
|
|
float loss = train_network_data_gpu(net, train, imgs);
|
2014-11-06 01:49:58 +03:00
|
|
|
avg_loss = avg_loss*.9 + loss*.1;
|
2014-11-22 02:35:19 +03:00
|
|
|
printf("%d: %f, %f avg, %lf seconds, %d images\n", i, loss, avg_loss, sec(clock()-time), i*imgs*net.batch);
|
|
|
|
#endif
|
|
|
|
free_data(train);
|
|
|
|
if(i%10==0){
|
|
|
|
char buff[256];
|
2014-12-03 19:48:07 +03:00
|
|
|
sprintf(buff, "/home/pjreddie/imagenet_backup/alexnet_%d.cfg", i);
|
2014-11-22 02:35:19 +03:00
|
|
|
save_network(net, buff);
|
|
|
|
}
|
|
|
|
}
|
2014-10-25 22:57:26 +04:00
|
|
|
}
|
|
|
|
|
2014-11-06 01:49:58 +03:00
|
|
|
void validate_imagenet(char *filename)
|
|
|
|
{
|
|
|
|
int i;
|
2014-11-22 02:35:19 +03:00
|
|
|
network net = parse_network_cfg(filename);
|
|
|
|
srand(time(0));
|
2014-11-06 01:49:58 +03:00
|
|
|
|
|
|
|
char **labels = get_labels("/home/pjreddie/data/imagenet/cls.val.labels.list");
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
list *plist = get_paths("/home/pjreddie/data/imagenet/cls.val.list");
|
|
|
|
char **paths = (char **)list_to_array(plist);
|
|
|
|
int m = plist->size;
|
|
|
|
free_list(plist);
|
2014-11-06 01:49:58 +03:00
|
|
|
|
|
|
|
clock_t time;
|
|
|
|
float avg_acc = 0;
|
|
|
|
int splits = 50;
|
2014-11-22 02:35:19 +03:00
|
|
|
|
2014-11-06 01:49:58 +03:00
|
|
|
for(i = 0; i < splits; ++i){
|
|
|
|
time=clock();
|
2014-11-22 02:35:19 +03:00
|
|
|
char **part = paths+(i*m/splits);
|
|
|
|
int num = (i+1)*m/splits - i*m/splits;
|
|
|
|
data val = load_data(part, num, labels, 1000, 256, 256);
|
2014-12-03 19:48:07 +03:00
|
|
|
|
2014-11-06 01:49:58 +03:00
|
|
|
normalize_data_rows(val);
|
|
|
|
printf("Loaded: %d images in %lf seconds\n", val.X.rows, sec(clock()-time));
|
|
|
|
time=clock();
|
2014-10-28 05:45:06 +03:00
|
|
|
#ifdef GPU
|
2014-11-22 02:35:19 +03:00
|
|
|
float acc = network_accuracy_gpu(net, val);
|
|
|
|
avg_acc += acc;
|
|
|
|
printf("%d: %f, %f avg, %lf seconds, %d images\n", i, acc, avg_acc/(i+1), sec(clock()-time), val.X.rows);
|
2014-10-28 05:45:06 +03:00
|
|
|
#endif
|
2014-11-22 02:35:19 +03:00
|
|
|
free_data(val);
|
|
|
|
}
|
2014-10-28 05:45:06 +03:00
|
|
|
}
|
|
|
|
|
2014-12-03 19:48:07 +03:00
|
|
|
void draw_detection(image im, float *box)
|
|
|
|
{
|
|
|
|
int j;
|
|
|
|
int r, c;
|
|
|
|
for(r = 0; r < 8; ++r){
|
|
|
|
for(c = 0; c < 8; ++c){
|
|
|
|
j = (r*8 + c) * 5;
|
|
|
|
printf("Prob: %f\n", box[j]);
|
|
|
|
if(box[j] > .999){
|
|
|
|
int d = 256/8;
|
|
|
|
int y = r*d+box[j+1]*d;
|
|
|
|
int x = c*d+box[j+2]*d;
|
|
|
|
int h = box[j+3]*256;
|
|
|
|
int w = box[j+4]*256;
|
|
|
|
printf("%f %f %f %f\n", box[j+1], box[j+2], box[j+3], box[j+4]);
|
|
|
|
printf("%d %d %d %d\n", x, y, w, h);
|
|
|
|
printf("%d %d %d %d\n", x-w/2, y-h/2, x+w/2, y+h/2);
|
|
|
|
draw_box(im, x-w/2, y-h/2, x+w/2, y+h/2);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
show_image(im, "box");
|
|
|
|
cvWaitKey(0);
|
|
|
|
}
|
|
|
|
|
2014-11-28 21:38:26 +03:00
|
|
|
void test_detection()
|
|
|
|
{
|
|
|
|
network net = parse_network_cfg("cfg/detnet_test.cfg");
|
|
|
|
srand(2222222);
|
|
|
|
clock_t time;
|
|
|
|
char filename[256];
|
|
|
|
while(1){
|
|
|
|
fgets(filename, 256, stdin);
|
2014-12-03 19:48:07 +03:00
|
|
|
strtok(filename, "\n");
|
2014-11-28 21:38:26 +03:00
|
|
|
image im = load_image_color(filename, 256, 256);
|
|
|
|
z_normalize_image(im);
|
|
|
|
printf("%d %d %d\n", im.h, im.w, im.c);
|
|
|
|
float *X = im.data;
|
|
|
|
time=clock();
|
|
|
|
float *predictions = network_predict(net, X);
|
|
|
|
printf("%s: Predicted in %f seconds.\n", filename, sec(clock()-time));
|
2014-12-03 19:48:07 +03:00
|
|
|
draw_detection(im, predictions);
|
2014-11-28 21:38:26 +03:00
|
|
|
free_image(im);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-10-25 22:57:26 +04:00
|
|
|
void test_imagenet()
|
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
network net = parse_network_cfg("cfg/imagenet_test.cfg");
|
2014-10-25 22:57:26 +04:00
|
|
|
//imgs=1;
|
2014-10-28 05:45:06 +03:00
|
|
|
srand(2222222);
|
|
|
|
int i = 0;
|
2014-10-25 22:57:26 +04:00
|
|
|
char **names = get_labels("cfg/shortnames.txt");
|
|
|
|
clock_t time;
|
|
|
|
char filename[256];
|
|
|
|
int indexes[10];
|
2014-10-28 05:45:06 +03:00
|
|
|
while(1){
|
2014-11-19 00:51:04 +03:00
|
|
|
fgets(filename, 256, stdin);
|
2014-12-03 19:48:07 +03:00
|
|
|
strtok(filename, "\n");
|
2014-10-25 22:57:26 +04:00
|
|
|
image im = load_image_color(filename, 256, 256);
|
2014-11-06 01:49:58 +03:00
|
|
|
z_normalize_image(im);
|
2014-10-25 22:57:26 +04:00
|
|
|
printf("%d %d %d\n", im.h, im.w, im.c);
|
|
|
|
float *X = im.data;
|
|
|
|
time=clock();
|
|
|
|
float *predictions = network_predict(net, X);
|
|
|
|
top_predictions(net, 10, indexes);
|
2014-10-28 05:45:06 +03:00
|
|
|
printf("%s: Predicted in %f seconds.\n", filename, sec(clock()-time));
|
2014-10-25 22:57:26 +04:00
|
|
|
for(i = 0; i < 10; ++i){
|
|
|
|
int index = indexes[i];
|
|
|
|
printf("%s: %f\n", names[index], predictions[index]);
|
|
|
|
}
|
2014-10-28 05:45:06 +03:00
|
|
|
free_image(im);
|
|
|
|
}
|
2014-10-25 22:57:26 +04:00
|
|
|
}
|
|
|
|
|
2014-11-06 01:49:58 +03:00
|
|
|
void test_visualize(char *filename)
|
2014-04-11 12:00:27 +04:00
|
|
|
{
|
2014-11-06 01:49:58 +03:00
|
|
|
network net = parse_network_cfg(filename);
|
2014-10-28 05:45:06 +03:00
|
|
|
visualize_network(net);
|
|
|
|
cvWaitKey(0);
|
2013-11-13 22:50:38 +04:00
|
|
|
}
|
|
|
|
|
2014-03-13 08:57:34 +04:00
|
|
|
void test_cifar10()
|
|
|
|
{
|
2014-08-11 23:52:07 +04:00
|
|
|
network net = parse_network_cfg("cfg/cifar10_part5.cfg");
|
|
|
|
data test = load_cifar10_data("data/cifar10/test_batch.bin");
|
2014-10-28 05:45:06 +03:00
|
|
|
clock_t start = clock(), end;
|
2014-08-11 23:52:07 +04:00
|
|
|
float test_acc = network_accuracy(net, test);
|
2014-10-28 05:45:06 +03:00
|
|
|
end = clock();
|
2014-08-11 23:52:07 +04:00
|
|
|
printf("%f in %f Sec\n", test_acc, (float)(end-start)/CLOCKS_PER_SEC);
|
|
|
|
visualize_network(net);
|
|
|
|
cvWaitKey(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
void train_cifar10()
|
|
|
|
{
|
|
|
|
srand(555555);
|
2014-08-08 23:04:15 +04:00
|
|
|
network net = parse_network_cfg("cfg/cifar10.cfg");
|
2014-08-11 23:52:07 +04:00
|
|
|
data test = load_cifar10_data("data/cifar10/test_batch.bin");
|
2014-08-08 23:04:15 +04:00
|
|
|
int count = 0;
|
|
|
|
int iters = 10000/net.batch;
|
|
|
|
data train = load_all_cifar10();
|
|
|
|
while(++count <= 10000){
|
|
|
|
clock_t start = clock(), end;
|
2014-10-22 01:49:18 +04:00
|
|
|
float loss = train_network_sgd(net, train, iters);
|
2014-08-08 23:04:15 +04:00
|
|
|
end = clock();
|
2014-10-17 02:17:23 +04:00
|
|
|
//visualize_network(net);
|
|
|
|
//cvWaitKey(5000);
|
2014-08-08 23:04:15 +04:00
|
|
|
|
|
|
|
//float test_acc = network_accuracy(net, test);
|
|
|
|
//printf("%d: Loss: %f, Test Acc: %f, Time: %lf seconds, LR: %f, Momentum: %f, Decay: %f\n", count, loss, test_acc,(float)(end-start)/CLOCKS_PER_SEC, net.learning_rate, net.momentum, net.decay);
|
2014-08-11 23:52:07 +04:00
|
|
|
if(count%10 == 0){
|
|
|
|
float test_acc = network_accuracy(net, test);
|
|
|
|
printf("%d: Loss: %f, Test Acc: %f, Time: %lf seconds, LR: %f, Momentum: %f, Decay: %f\n", count, loss, test_acc,(float)(end-start)/CLOCKS_PER_SEC, net.learning_rate, net.momentum, net.decay);
|
|
|
|
char buff[256];
|
2014-10-22 01:49:18 +04:00
|
|
|
sprintf(buff, "/home/pjreddie/cifar/cifar10_2_%d.cfg", count);
|
2014-08-11 23:52:07 +04:00
|
|
|
save_network(net, buff);
|
|
|
|
}else{
|
|
|
|
printf("%d: Loss: %f, Time: %lf seconds, LR: %f, Momentum: %f, Decay: %f\n", count, loss, (float)(end-start)/CLOCKS_PER_SEC, net.learning_rate, net.momentum, net.decay);
|
|
|
|
}
|
2014-08-08 23:04:15 +04:00
|
|
|
}
|
|
|
|
free_data(train);
|
2014-03-13 08:57:34 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
void test_vince()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
network net = parse_network_cfg("cfg/vince.cfg");
|
|
|
|
data train = load_categorical_data_csv("images/vince.txt", 144, 2);
|
|
|
|
normalize_data_rows(train);
|
|
|
|
|
|
|
|
int count = 0;
|
|
|
|
//float lr = .00005;
|
|
|
|
//float momentum = .9;
|
|
|
|
//float decay = 0.0001;
|
|
|
|
//decay = 0;
|
|
|
|
int batch = 10000;
|
|
|
|
while(++count <= 10000){
|
|
|
|
float loss = train_network_sgd(net, train, batch);
|
|
|
|
printf("%5f %5f\n",(double)count*batch/train.X.rows, loss);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void test_nist_single()
|
|
|
|
{
|
|
|
|
srand(222222);
|
2014-08-09 19:16:37 +04:00
|
|
|
network net = parse_network_cfg("cfg/nist_single.cfg");
|
2014-08-08 23:04:15 +04:00
|
|
|
data train = load_categorical_data_csv("data/mnist/mnist_tiny.csv", 0, 10);
|
|
|
|
normalize_data_rows(train);
|
2014-08-09 19:16:37 +04:00
|
|
|
float loss = train_network_sgd(net, train, 1);
|
2014-08-08 23:04:15 +04:00
|
|
|
printf("Loss: %f, LR: %f, Momentum: %f, Decay: %f\n", loss, net.learning_rate, net.momentum, net.decay);
|
2014-04-17 20:57:30 +04:00
|
|
|
|
2014-02-25 00:21:31 +04:00
|
|
|
}
|
2013-11-13 22:50:38 +04:00
|
|
|
|
2013-12-03 04:41:40 +04:00
|
|
|
void test_nist()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
srand(222222);
|
2014-08-11 23:52:07 +04:00
|
|
|
network net = parse_network_cfg("cfg/nist_final.cfg");
|
|
|
|
data test = load_categorical_data_csv("data/mnist/mnist_test.csv",0,10);
|
|
|
|
translate_data_rows(test, -144);
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
float test_acc = network_accuracy_multi(net, test,16);
|
|
|
|
end = clock();
|
|
|
|
printf("Accuracy: %f, Time: %lf seconds\n", test_acc,(float)(end-start)/CLOCKS_PER_SEC);
|
|
|
|
}
|
|
|
|
|
|
|
|
void train_nist()
|
|
|
|
{
|
|
|
|
srand(222222);
|
2014-08-28 06:11:46 +04:00
|
|
|
network net = parse_network_cfg("cfg/nist.cfg");
|
2014-08-08 23:04:15 +04:00
|
|
|
data train = load_categorical_data_csv("data/mnist/mnist_train.csv", 0, 10);
|
|
|
|
data test = load_categorical_data_csv("data/mnist/mnist_test.csv",0,10);
|
2014-08-11 23:52:07 +04:00
|
|
|
translate_data_rows(train, -144);
|
|
|
|
translate_data_rows(test, -144);
|
2014-08-08 23:04:15 +04:00
|
|
|
int count = 0;
|
2014-11-19 00:51:04 +03:00
|
|
|
int iters = 50000/net.batch;
|
2014-08-09 19:16:37 +04:00
|
|
|
while(++count <= 2000){
|
2014-08-08 23:04:15 +04:00
|
|
|
clock_t start = clock(), end;
|
2014-10-28 05:45:06 +03:00
|
|
|
float loss = train_network_sgd(net, train, iters);
|
2014-08-08 23:04:15 +04:00
|
|
|
end = clock();
|
|
|
|
float test_acc = network_accuracy(net, test);
|
2014-11-19 00:51:04 +03:00
|
|
|
printf("%d: Loss: %f, Test Acc: %f, Time: %lf seconds\n", count, loss, test_acc,(float)(end-start)/CLOCKS_PER_SEC);
|
2014-08-08 23:04:15 +04:00
|
|
|
}
|
2013-12-03 04:41:40 +04:00
|
|
|
}
|
|
|
|
|
2013-12-07 21:38:50 +04:00
|
|
|
void test_ensemble()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
int i;
|
|
|
|
srand(888888);
|
|
|
|
data d = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
|
|
|
|
normalize_data_rows(d);
|
|
|
|
data test = load_categorical_data_csv("mnist/mnist_test.csv", 0,10);
|
|
|
|
normalize_data_rows(test);
|
|
|
|
data train = d;
|
|
|
|
// data *split = split_data(d, 1, 10);
|
|
|
|
// data train = split[0];
|
|
|
|
// data test = split[1];
|
|
|
|
matrix prediction = make_matrix(test.y.rows, test.y.cols);
|
|
|
|
int n = 30;
|
|
|
|
for(i = 0; i < n; ++i){
|
|
|
|
int count = 0;
|
|
|
|
float lr = .0005;
|
|
|
|
float momentum = .9;
|
|
|
|
float decay = .01;
|
|
|
|
network net = parse_network_cfg("nist.cfg");
|
|
|
|
while(++count <= 15){
|
|
|
|
float acc = train_network_sgd(net, train, train.X.rows);
|
|
|
|
printf("Training Accuracy: %lf Learning Rate: %f Momentum: %f Decay: %f\n", acc, lr, momentum, decay );
|
|
|
|
lr /= 2;
|
|
|
|
}
|
|
|
|
matrix partial = network_predict_data(net, test);
|
|
|
|
float acc = matrix_accuracy(test.y, partial);
|
|
|
|
printf("Model Accuracy: %lf\n", acc);
|
|
|
|
matrix_add_matrix(partial, prediction);
|
|
|
|
acc = matrix_accuracy(test.y, prediction);
|
|
|
|
printf("Current Ensemble Accuracy: %lf\n", acc);
|
|
|
|
free_matrix(partial);
|
|
|
|
}
|
|
|
|
float acc = matrix_accuracy(test.y, prediction);
|
|
|
|
printf("Full Ensemble Accuracy: %lf\n", acc);
|
2013-12-07 21:38:50 +04:00
|
|
|
}
|
|
|
|
|
2013-12-03 04:41:40 +04:00
|
|
|
void test_random_classify()
|
2013-11-13 22:50:38 +04:00
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
network net = parse_network_cfg("connected.cfg");
|
|
|
|
matrix m = csv_to_matrix("train.csv");
|
|
|
|
//matrix ho = hold_out_matrix(&m, 2500);
|
|
|
|
float *truth = pop_column(&m, 0);
|
|
|
|
//float *ho_truth = pop_column(&ho, 0);
|
|
|
|
int i;
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
int count = 0;
|
|
|
|
while(++count <= 300){
|
|
|
|
for(i = 0; i < m.rows; ++i){
|
|
|
|
int index = rand()%m.rows;
|
|
|
|
//image p = float_to_image(1690,1,1,m.vals[index]);
|
|
|
|
//normalize_image(p);
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network(net, m.vals[index], 0, 1);
|
2014-08-08 23:04:15 +04:00
|
|
|
float *out = get_network_output(net);
|
|
|
|
float *delta = get_network_delta(net);
|
|
|
|
//printf("%f\n", out[0]);
|
|
|
|
delta[0] = truth[index] - out[0];
|
|
|
|
// printf("%f\n", delta[0]);
|
|
|
|
//printf("%f %f\n", truth[index], out[0]);
|
|
|
|
//backward_network(net, m.vals[index], );
|
|
|
|
update_network(net);
|
|
|
|
}
|
|
|
|
//float test_acc = error_network(net, m, truth);
|
|
|
|
//float valid_acc = error_network(net, ho, ho_truth);
|
|
|
|
//printf("%f, %f\n", test_acc, valid_acc);
|
|
|
|
//fprintf(stderr, "%5d: %f Valid: %f\n",count, test_acc, valid_acc);
|
|
|
|
//if(valid_acc > .70) break;
|
|
|
|
}
|
|
|
|
end = clock();
|
|
|
|
FILE *fp = fopen("submission/out.txt", "w");
|
|
|
|
matrix test = csv_to_matrix("test.csv");
|
|
|
|
truth = pop_column(&test, 0);
|
|
|
|
for(i = 0; i < test.rows; ++i){
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network(net, test.vals[i],0, 0);
|
2014-08-08 23:04:15 +04:00
|
|
|
float *out = get_network_output(net);
|
|
|
|
if(fabs(out[0]) < .5) fprintf(fp, "0\n");
|
|
|
|
else fprintf(fp, "1\n");
|
|
|
|
}
|
|
|
|
fclose(fp);
|
|
|
|
printf("Neural Net Learning: %lf seconds\n", (float)(end-start)/CLOCKS_PER_SEC);
|
2013-11-13 22:50:38 +04:00
|
|
|
}
|
|
|
|
|
2013-12-07 01:26:09 +04:00
|
|
|
void test_split()
|
2013-11-13 22:50:38 +04:00
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
data train = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
|
|
|
|
data *split = split_data(train, 0, 13);
|
|
|
|
printf("%d, %d, %d\n", train.X.rows, split[0].X.rows, split[1].X.rows);
|
2013-11-07 04:09:41 +04:00
|
|
|
}
|
|
|
|
|
2014-01-25 02:49:02 +04:00
|
|
|
void test_im2row()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
int h = 20;
|
|
|
|
int w = 20;
|
|
|
|
int c = 3;
|
|
|
|
int stride = 1;
|
|
|
|
int size = 11;
|
|
|
|
image test = make_random_image(h,w,c);
|
|
|
|
int mc = 1;
|
|
|
|
int mw = ((h-size)/stride+1)*((w-size)/stride+1);
|
|
|
|
int mh = (size*size*c);
|
|
|
|
int msize = mc*mw*mh;
|
|
|
|
float *matrix = calloc(msize, sizeof(float));
|
|
|
|
int i;
|
|
|
|
for(i = 0; i < 1000; ++i){
|
|
|
|
im2col_cpu(test.data,1, c, h, w, size, stride, 0, matrix);
|
|
|
|
//image render = float_to_image(mh, mw, mc, matrix);
|
|
|
|
}
|
2014-02-14 22:26:31 +04:00
|
|
|
}
|
|
|
|
|
2014-04-17 04:05:29 +04:00
|
|
|
void flip_network()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
network net = parse_network_cfg("cfg/voc_imagenet_orig.cfg");
|
|
|
|
save_network(net, "cfg/voc_imagenet_rev.cfg");
|
2014-02-14 22:26:31 +04:00
|
|
|
}
|
|
|
|
|
2014-02-18 11:32:18 +04:00
|
|
|
|
2014-11-22 02:35:19 +03:00
|
|
|
void visualize_cat()
|
2014-02-18 11:32:18 +04:00
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
network net = parse_network_cfg("cfg/voc_imagenet.cfg");
|
|
|
|
image im = load_image("data/cat.png", 0, 0);
|
|
|
|
printf("Processing %dx%d image\n", im.h, im.w);
|
2014-08-08 23:04:15 +04:00
|
|
|
resize_network(net, im.h, im.w, im.c);
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network(net, im.data, 0, 0);
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
visualize_network(net);
|
|
|
|
cvWaitKey(0);
|
2014-02-18 11:32:18 +04:00
|
|
|
}
|
|
|
|
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
void test_gpu_net()
|
2014-02-18 11:32:18 +04:00
|
|
|
{
|
2014-11-22 02:35:19 +03:00
|
|
|
srand(222222);
|
|
|
|
network net = parse_network_cfg("cfg/nist.cfg");
|
|
|
|
data train = load_categorical_data_csv("data/mnist/mnist_train.csv", 0, 10);
|
|
|
|
data test = load_categorical_data_csv("data/mnist/mnist_test.csv",0,10);
|
|
|
|
translate_data_rows(train, -144);
|
|
|
|
translate_data_rows(test, -144);
|
|
|
|
int count = 0;
|
|
|
|
int iters = 1000/net.batch;
|
|
|
|
while(++count <= 5){
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
float loss = train_network_sgd(net, train, iters);
|
|
|
|
end = clock();
|
|
|
|
float test_acc = network_accuracy(net, test);
|
|
|
|
printf("%d: Loss: %f, Test Acc: %f, Time: %lf seconds, LR: %f, Momentum: %f, Decay: %f\n", count, loss, test_acc,(float)(end-start)/CLOCKS_PER_SEC, net.learning_rate, net.momentum, net.decay);
|
|
|
|
}
|
|
|
|
#ifdef GPU
|
|
|
|
count = 0;
|
|
|
|
srand(222222);
|
|
|
|
net = parse_network_cfg("cfg/nist.cfg");
|
|
|
|
while(++count <= 5){
|
|
|
|
clock_t start = clock(), end;
|
|
|
|
float loss = train_network_sgd_gpu(net, train, iters);
|
|
|
|
end = clock();
|
|
|
|
float test_acc = network_accuracy(net, test);
|
|
|
|
printf("%d: Loss: %f, Test Acc: %f, Time: %lf seconds, LR: %f, Momentum: %f, Decay: %f\n", count, loss, test_acc,(float)(end-start)/CLOCKS_PER_SEC, net.learning_rate, net.momentum, net.decay);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
}
|
2014-08-08 23:04:15 +04:00
|
|
|
|
2014-12-03 19:48:07 +03:00
|
|
|
void test_server()
|
|
|
|
{
|
|
|
|
server_update();
|
|
|
|
}
|
|
|
|
void test_client()
|
|
|
|
{
|
|
|
|
client_update();
|
|
|
|
}
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
int main(int argc, char *argv[])
|
|
|
|
{
|
|
|
|
if(argc < 2){
|
|
|
|
fprintf(stderr, "usage: %s <function>\n", argv[0]);
|
|
|
|
return 0;
|
2014-08-08 23:04:15 +04:00
|
|
|
}
|
2014-11-22 02:35:19 +03:00
|
|
|
if(0==strcmp(argv[1], "train")) train_imagenet();
|
2014-11-28 21:38:26 +03:00
|
|
|
else if(0==strcmp(argv[1], "detection")) train_detection_net();
|
2014-11-22 02:35:19 +03:00
|
|
|
else if(0==strcmp(argv[1], "asirra")) train_asirra();
|
|
|
|
else if(0==strcmp(argv[1], "nist")) train_nist();
|
|
|
|
else if(0==strcmp(argv[1], "test_correct")) test_gpu_net();
|
|
|
|
else if(0==strcmp(argv[1], "test")) test_imagenet();
|
2014-12-03 19:48:07 +03:00
|
|
|
else if(0==strcmp(argv[1], "server")) test_server();
|
|
|
|
else if(0==strcmp(argv[1], "client")) test_client();
|
|
|
|
else if(0==strcmp(argv[1], "detect")) test_detection();
|
2014-11-22 02:35:19 +03:00
|
|
|
else if(0==strcmp(argv[1], "visualize")) test_visualize(argv[2]);
|
|
|
|
else if(0==strcmp(argv[1], "valid")) validate_imagenet(argv[2]);
|
|
|
|
#ifdef GPU
|
|
|
|
else if(0==strcmp(argv[1], "test_gpu")) test_gpu_blas();
|
|
|
|
#endif
|
|
|
|
fprintf(stderr, "Success!\n");
|
|
|
|
return 0;
|
2014-02-15 04:09:07 +04:00
|
|
|
}
|
2014-11-22 02:35:19 +03:00
|
|
|
|
|
|
|
/*
|
2014-12-03 19:48:07 +03:00
|
|
|
void visualize_imagenet_topk(char *filename)
|
|
|
|
{
|
|
|
|
int i,j,k,l;
|
|
|
|
int topk = 10;
|
|
|
|
network net = parse_network_cfg("cfg/voc_imagenet.cfg");
|
|
|
|
list *plist = get_paths(filename);
|
|
|
|
node *n = plist->front;
|
|
|
|
int h = voc_size(1), w = voc_size(1);
|
|
|
|
int num = get_network_image(net).c;
|
|
|
|
image **vizs = calloc(num, sizeof(image*));
|
|
|
|
float **score = calloc(num, sizeof(float *));
|
|
|
|
for(i = 0; i < num; ++i){
|
|
|
|
vizs[i] = calloc(topk, sizeof(image));
|
|
|
|
for(j = 0; j < topk; ++j) vizs[i][j] = make_image(h,w,3);
|
|
|
|
score[i] = calloc(topk, sizeof(float));
|
|
|
|
}
|
|
|
|
|
|
|
|
int count = 0;
|
|
|
|
while(n){
|
|
|
|
++count;
|
|
|
|
char *image_path = (char *)n->val;
|
|
|
|
image im = load_image(image_path, 0, 0);
|
|
|
|
n = n->next;
|
|
|
|
if(im.h < 200 || im.w < 200) continue;
|
|
|
|
printf("Processing %dx%d image\n", im.h, im.w);
|
|
|
|
resize_network(net, im.h, im.w, im.c);
|
|
|
|
//scale_image(im, 1./255);
|
|
|
|
translate_image(im, -144);
|
|
|
|
forward_network(net, im.data, 0, 0);
|
|
|
|
image out = get_network_image(net);
|
|
|
|
|
|
|
|
int dh = (im.h - h)/(out.h-1);
|
|
|
|
int dw = (im.w - w)/(out.w-1);
|
|
|
|
//printf("%d %d\n", dh, dw);
|
|
|
|
for(k = 0; k < out.c; ++k){
|
|
|
|
float topv = 0;
|
|
|
|
int topi = -1;
|
|
|
|
int topj = -1;
|
|
|
|
for(i = 0; i < out.h; ++i){
|
|
|
|
for(j = 0; j < out.w; ++j){
|
|
|
|
float val = get_pixel(out, i, j, k);
|
|
|
|
if(val > topv){
|
|
|
|
topv = val;
|
|
|
|
topi = i;
|
|
|
|
topj = j;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if(topv){
|
|
|
|
image sub = get_sub_image(im, dh*topi, dw*topj, h, w);
|
|
|
|
for(l = 0; l < topk; ++l){
|
|
|
|
if(topv > score[k][l]){
|
|
|
|
float swap = score[k][l];
|
|
|
|
score[k][l] = topv;
|
|
|
|
topv = swap;
|
|
|
|
|
|
|
|
image swapi = vizs[k][l];
|
|
|
|
vizs[k][l] = sub;
|
|
|
|
sub = swapi;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
free_image(sub);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
free_image(im);
|
|
|
|
if(count%50 == 0){
|
|
|
|
image grid = grid_images(vizs, num, topk);
|
|
|
|
//show_image(grid, "IMAGENET Visualization");
|
|
|
|
save_image(grid, "IMAGENET Grid Single Nonorm");
|
|
|
|
free_image(grid);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
//cvWaitKey(0);
|
2014-04-17 04:05:29 +04:00
|
|
|
}
|
2014-02-15 04:09:07 +04:00
|
|
|
|
2014-04-11 12:00:27 +04:00
|
|
|
void visualize_imagenet_features(char *filename)
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
int i,j,k;
|
|
|
|
network net = parse_network_cfg("cfg/voc_imagenet.cfg");
|
|
|
|
list *plist = get_paths(filename);
|
|
|
|
node *n = plist->front;
|
|
|
|
int h = voc_size(1), w = voc_size(1);
|
|
|
|
int num = get_network_image(net).c;
|
|
|
|
image *vizs = calloc(num, sizeof(image));
|
|
|
|
for(i = 0; i < num; ++i) vizs[i] = make_image(h, w, 3);
|
|
|
|
while(n){
|
|
|
|
char *image_path = (char *)n->val;
|
|
|
|
image im = load_image(image_path, 0, 0);
|
|
|
|
printf("Processing %dx%d image\n", im.h, im.w);
|
|
|
|
resize_network(net, im.h, im.w, im.c);
|
2014-10-13 11:29:01 +04:00
|
|
|
forward_network(net, im.data, 0, 0);
|
2014-08-08 23:04:15 +04:00
|
|
|
image out = get_network_image(net);
|
|
|
|
|
|
|
|
int dh = (im.h - h)/h;
|
|
|
|
int dw = (im.w - w)/w;
|
|
|
|
for(i = 0; i < out.h; ++i){
|
|
|
|
for(j = 0; j < out.w; ++j){
|
|
|
|
image sub = get_sub_image(im, dh*i, dw*j, h, w);
|
|
|
|
for(k = 0; k < out.c; ++k){
|
|
|
|
float val = get_pixel(out, i, j, k);
|
|
|
|
//printf("%f, ", val);
|
|
|
|
image sub_c = copy_image(sub);
|
|
|
|
scale_image(sub_c, val);
|
|
|
|
add_into_image(sub_c, vizs[k], 0, 0);
|
|
|
|
free_image(sub_c);
|
|
|
|
}
|
|
|
|
free_image(sub);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
//printf("\n");
|
|
|
|
show_images(vizs, 10, "IMAGENET Visualization");
|
|
|
|
cvWaitKey(1000);
|
|
|
|
n = n->next;
|
|
|
|
}
|
|
|
|
cvWaitKey(0);
|
2014-02-18 11:32:18 +04:00
|
|
|
}
|
2014-05-10 02:35:58 +04:00
|
|
|
void features_VOC_image(char *image_file, char *image_dir, char *out_dir, int flip, int interval)
|
2014-02-20 01:41:44 +04:00
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
int i,j;
|
|
|
|
network net = parse_network_cfg("cfg/voc_imagenet.cfg");
|
|
|
|
char image_path[1024];
|
|
|
|
sprintf(image_path, "%s/%s",image_dir, image_file);
|
|
|
|
char out_path[1024];
|
|
|
|
if (flip)sprintf(out_path, "%s%d/%s_r.txt",out_dir, interval, image_file);
|
|
|
|
else sprintf(out_path, "%s%d/%s.txt",out_dir, interval, image_file);
|
|
|
|
printf("%s\n", image_file);
|
|
|
|
|
|
|
|
IplImage* src = 0;
|
|
|
|
if( (src = cvLoadImage(image_path,-1)) == 0 ) file_error(image_path);
|
|
|
|
if(flip)cvFlip(src, 0, 1);
|
|
|
|
int w = src->width;
|
|
|
|
int h = src->height;
|
|
|
|
int sbin = 8;
|
|
|
|
double scale = pow(2., 1./interval);
|
|
|
|
int m = (w<h)?w:h;
|
|
|
|
int max_scale = 1+floor((double)log((double)m/(5.*sbin))/log(scale));
|
|
|
|
if(max_scale < interval) error("max_scale must be >= interval");
|
|
|
|
image *ims = calloc(max_scale+interval, sizeof(image));
|
|
|
|
|
|
|
|
for(i = 0; i < interval; ++i){
|
|
|
|
double factor = 1./pow(scale, i);
|
|
|
|
double ih = round(h*factor);
|
|
|
|
double iw = round(w*factor);
|
|
|
|
int ex_h = round(ih/4.) - 2;
|
|
|
|
int ex_w = round(iw/4.) - 2;
|
|
|
|
ims[i] = features_output_size(net, src, ex_h, ex_w);
|
|
|
|
|
|
|
|
ih = round(h*factor);
|
|
|
|
iw = round(w*factor);
|
|
|
|
ex_h = round(ih/8.) - 2;
|
|
|
|
ex_w = round(iw/8.) - 2;
|
|
|
|
ims[i+interval] = features_output_size(net, src, ex_h, ex_w);
|
|
|
|
for(j = i+interval; j < max_scale; j += interval){
|
|
|
|
factor /= 2.;
|
|
|
|
ih = round(h*factor);
|
|
|
|
iw = round(w*factor);
|
|
|
|
ex_h = round(ih/8.) - 2;
|
|
|
|
ex_w = round(iw/8.) - 2;
|
|
|
|
ims[j+interval] = features_output_size(net, src, ex_h, ex_w);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
FILE *fp = fopen(out_path, "w");
|
|
|
|
if(fp == 0) file_error(out_path);
|
|
|
|
for(i = 0; i < max_scale+interval; ++i){
|
|
|
|
image out = ims[i];
|
|
|
|
fprintf(fp, "%d, %d, %d\n",out.c, out.h, out.w);
|
|
|
|
for(j = 0; j < out.c*out.h*out.w; ++j){
|
|
|
|
if(j != 0)fprintf(fp, ",");
|
|
|
|
float o = out.data[j];
|
|
|
|
if(o < 0) o = 0;
|
|
|
|
fprintf(fp, "%g", o);
|
|
|
|
}
|
|
|
|
fprintf(fp, "\n");
|
|
|
|
free_image(out);
|
|
|
|
}
|
|
|
|
free(ims);
|
|
|
|
fclose(fp);
|
|
|
|
cvReleaseImage(&src);
|
2014-02-20 01:41:44 +04:00
|
|
|
}
|
|
|
|
|
2014-02-25 00:21:31 +04:00
|
|
|
void test_distribution()
|
|
|
|
{
|
2014-08-08 23:04:15 +04:00
|
|
|
IplImage* img = 0;
|
|
|
|
if( (img = cvLoadImage("im_small.jpg",-1)) == 0 ) file_error("im_small.jpg");
|
|
|
|
network net = parse_network_cfg("cfg/voc_features.cfg");
|
|
|
|
int h = img->height/8-2;
|
|
|
|
int w = img->width/8-2;
|
|
|
|
image out = features_output_size(net, img, h, w);
|
|
|
|
int c = out.c;
|
|
|
|
out.c = 1;
|
|
|
|
show_image(out, "output");
|
|
|
|
out.c = c;
|
|
|
|
image input = ipl_to_image(img);
|
|
|
|
show_image(input, "input");
|
|
|
|
CvScalar s;
|
|
|
|
int i,j;
|
|
|
|
image affects = make_image(input.h, input.w, 1);
|
|
|
|
int count = 0;
|
|
|
|
for(i = 0; i<img->height; i += 1){
|
|
|
|
for(j = 0; j < img->width; j += 1){
|
|
|
|
IplImage *copy = cvCloneImage(img);
|
|
|
|
s=cvGet2D(copy,i,j); // get the (i,j) pixel value
|
|
|
|
printf("%d/%d\n", count++, img->height*img->width);
|
|
|
|
s.val[0]=0;
|
|
|
|
s.val[1]=0;
|
|
|
|
s.val[2]=0;
|
|
|
|
cvSet2D(copy,i,j,s); // set the (i,j) pixel value
|
|
|
|
image mod = features_output_size(net, copy, h, w);
|
|
|
|
image dist = image_distance(out, mod);
|
|
|
|
show_image(affects, "affects");
|
|
|
|
cvWaitKey(1);
|
|
|
|
cvReleaseImage(©);
|
|
|
|
//affects.data[i*affects.w + j] += dist.data[3*dist.w+5];
|
|
|
|
affects.data[i*affects.w + j] += dist.data[1*dist.w+1];
|
|
|
|
free_image(mod);
|
|
|
|
free_image(dist);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
show_image(affects, "Origins");
|
|
|
|
cvWaitKey(0);
|
|
|
|
cvWaitKey(0);
|
2014-02-25 00:21:31 +04:00
|
|
|
}
|
2014-11-22 02:35:19 +03:00
|
|
|
*/
|