From 4bdf96bd6aafbec6bc3f0eab8739d6652878fd24 Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Fri, 06 Dec 2013 21:26:09 +0000
Subject: [PATCH] New data format
---
src/network.c | 97 +++++++---
src/matrix.c | 11 +
src/utils.h | 1
src/network.h | 11
Makefile | 8
src/convolutional_layer.c | 13 -
src/connected_layer.c | 1
src/parser.c | 2
src/data.c | 179 +++++++++++++------
src/data.h | 24 +-
src/tests.c | 162 ++++--------------
src/utils.c | 13 +
12 files changed, 279 insertions(+), 243 deletions(-)
diff --git a/Makefile b/Makefile
index 6cd3999..e1238d6 100644
--- a/Makefile
+++ b/Makefile
@@ -1,5 +1,11 @@
CC=gcc
-COMMON=-Wall `pkg-config --cflags opencv` -isystem /usr/local/Cellar/opencv/2.4.6.1/include/opencv -isystem /usr/local/Cellar/opencv/2.4.6.1/include
+COMMON=-Wall `pkg-config --cflags opencv`
+UNAME = $(shell uname)
+ifeq ($(UNAME), Darwin)
+COMMON += -isystem /usr/local/Cellar/opencv/2.4.6.1/include/opencv -isystem /usr/local/Cellar/opencv/2.4.6.1/include
+else
+COMMON += -march=native
+endif
CFLAGS= $(COMMON) -O3 -ffast-math -flto
#CFLAGS= $(COMMON) -O0 -g
LDFLAGS=`pkg-config --libs opencv` -lm
diff --git a/src/connected_layer.c b/src/connected_layer.c
index d769e1f..0344c71 100644
--- a/src/connected_layer.c
+++ b/src/connected_layer.c
@@ -69,7 +69,6 @@
int index = i*layer.inputs+j;
layer.weight_momentum[index] = step*(layer.weight_updates[index] - decay*layer.weights[index]) + momentum*layer.weight_momentum[index];
layer.weights[index] += layer.weight_momentum[index];
- //layer.weights[index] = constrain(layer.weights[index], 100.);
}
}
memset(layer.bias_updates, 0, layer.outputs*sizeof(double));
diff --git a/src/convolutional_layer.c b/src/convolutional_layer.c
index 45b55b8..5accaab 100644
--- a/src/convolutional_layer.c
+++ b/src/convolutional_layer.c
@@ -143,26 +143,22 @@
for(i = 0; i < layer.n; ++i){
kernel_update(in_image, layer.kernel_updates[i], layer.stride, i, out_delta, layer.edge);
layer.bias_updates[i] += avg_image_layer(out_delta, i);
- //printf("%30.20lf\n", layer.bias_updates[i]);
}
}
void update_convolutional_layer(convolutional_layer layer, double step, double momentum, double decay)
{
- //step = .01;
int i,j;
for(i = 0; i < layer.n; ++i){
layer.bias_momentum[i] = step*(layer.bias_updates[i])
+ momentum*layer.bias_momentum[i];
layer.biases[i] += layer.bias_momentum[i];
- //layer.biases[i] = constrain(layer.biases[i],1.);
layer.bias_updates[i] = 0;
int pixels = layer.kernels[i].h*layer.kernels[i].w*layer.kernels[i].c;
for(j = 0; j < pixels; ++j){
layer.kernel_momentum[i].data[j] = step*(layer.kernel_updates[i].data[j] - decay*layer.kernels[i].data[j])
+ momentum*layer.kernel_momentum[i].data[j];
layer.kernels[i].data[j] += layer.kernel_momentum[i].data[j];
- //layer.kernels[i].data[j] = constrain(layer.kernels[i].data[j], 1.);
}
zero_image(layer.kernel_updates[i]);
}
@@ -188,14 +184,6 @@
int w_offset = i*(size+border);
image k = layer.kernels[i];
image copy = copy_image(k);
- /*
- printf("Kernel %d - Bias: %f, Channels:",i,layer.biases[i]);
- for(j = 0; j < k.c; ++j){
- double a = avg_image_layer(k, j);
- printf("%f, ", a);
- }
- printf("\n");
- */
normalize_image(copy);
for(j = 0; j < k.c; ++j){
set_pixel(copy,0,0,j,layer.biases[i]);
@@ -227,7 +215,6 @@
{
int i;
char buff[256];
- //image vis = make_image(layer.n*layer.size, layer.size*layer.kernels[0].c, 3);
for(i = 0; i < layer.n; ++i){
image k = layer.kernels[i];
sprintf(buff, "Kernel %d", i);
diff --git a/src/data.c b/src/data.c
index 9e5791f..b209197 100644
--- a/src/data.c
+++ b/src/data.c
@@ -1,23 +1,12 @@
#include "data.h"
#include "list.h"
#include "utils.h"
+#include "image.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
-batch make_batch(int n, int k)
-{
- batch b;
- b.n = n;
- if(k < 3) k = 1;
- b.images = calloc(n, sizeof(image));
- b.truth = calloc(n, sizeof(double *));
- int i;
- for(i =0 ; i < n; ++i) b.truth[i] = calloc(k, sizeof(double));
- return b;
-}
-
list *get_paths(char *filename)
{
char *path;
@@ -41,75 +30,145 @@
}
}
-batch load_list(list *paths, char **labels, int k)
-{
- char *path;
- batch data = make_batch(paths->size, 2);
- node *n = paths->front;
- int i;
- for(i = 0; i < data.n; ++i){
- path = (char *)n->val;
- data.images[i] = load_image(path);
- fill_truth(path, labels, k, data.truth[i]);
- n = n->next;
- }
- return data;
-}
-
-batch get_all_data(char *filename, char **labels, int k)
-{
- list *paths = get_paths(filename);
- batch b = load_list(paths, labels, k);
- free_list_contents(paths);
- free_list(paths);
- return b;
-}
-
-void free_batch(batch b)
+data load_data_image_paths(char **paths, int n, char **labels, int k)
{
int i;
- for(i = 0; i < b.n; ++i){
- free_image(b.images[i]);
- free(b.truth[i]);
+ data d;
+ d.shallow = 0;
+ d.X.rows = n;
+ d.X.vals = calloc(d.X.rows, sizeof(double*));
+ d.y = make_matrix(n, k);
+
+ for(i = 0; i < n; ++i){
+ image im = load_image(paths[i]);
+ d.X.vals[i] = im.data;
+ d.X.cols = im.h*im.w*im.c;
+ fill_truth(paths[i], labels, k, d.y.vals[i]);
}
- free(b.images);
- free(b.truth);
+ return d;
}
-batch get_batch(char *filename, int curr, int total, char **labels, int k)
+data load_data_image_pathfile(char *filename, char **labels, int k)
{
list *plist = get_paths(filename);
char **paths = (char **)list_to_array(plist);
- int i;
- int start = curr*plist->size/total;
- int end = (curr+1)*plist->size/total;
- batch b = make_batch(end-start, 2);
- for(i = start; i < end; ++i){
- b.images[i-start] = load_image(paths[i]);
- fill_truth(paths[i], labels, k, b.truth[i-start]);
- }
+ data d = load_data_image_paths(paths, plist->size, labels, k);
free_list_contents(plist);
free_list(plist);
free(paths);
- return b;
+ return d;
}
-batch random_batch(char *filename, int n, char **labels, int k)
+void free_data(data d)
+{
+ if(!d.shallow){
+ free_matrix(d.X);
+ free_matrix(d.y);
+ }else{
+ free(d.X.vals);
+ free(d.y.vals);
+ }
+}
+
+data load_data_image_pathfile_part(char *filename, int part, int total, char **labels, int k)
{
list *plist = get_paths(filename);
char **paths = (char **)list_to_array(plist);
+ int start = part*plist->size/total;
+ int end = (part+1)*plist->size/total;
+ data d = load_data_image_paths(paths+start, end-start, labels, k);
+ free_list_contents(plist);
+ free_list(plist);
+ free(paths);
+ return d;
+}
+
+data load_data_image_pathfile_random(char *filename, int n, char **labels, int k)
+{
int i;
- batch b = make_batch(n, 2);
+ list *plist = get_paths(filename);
+ char **paths = (char **)list_to_array(plist);
+ char **random_paths = calloc(n, sizeof(char*));
for(i = 0; i < n; ++i){
int index = rand()%plist->size;
- b.images[i] = load_image(paths[index]);
- //scale_image(b.images[i], 1./255.);
- z_normalize_image(b.images[i]);
- fill_truth(paths[index], labels, k, b.truth[i]);
- //print_image(b.images[i]);
+ random_paths[i] = paths[index];
}
+ data d = load_data_image_paths(random_paths, n, labels, k);
free_list_contents(plist);
free_list(plist);
free(paths);
- return b;
+ free(random_paths);
+ return d;
}
+
+data load_categorical_data_csv(char *filename, int target, int k)
+{
+ data d;
+ d.shallow = 0;
+ matrix X = csv_to_matrix(filename);
+ double *truth_1d = pop_column(&X, target);
+ double **truth = one_hot_encode(truth_1d, X.rows, k);
+ matrix y;
+ y.rows = X.rows;
+ y.cols = k;
+ y.vals = truth;
+ d.X = X;
+ d.y = y;
+ free(truth_1d);
+ return d;
+}
+
+void randomize_data(data d)
+{
+ int i;
+ for(i = d.X.rows-1; i > 0; --i){
+ int index = rand()%i;
+ double *swap = d.X.vals[index];
+ d.X.vals[index] = d.X.vals[i];
+ d.X.vals[i] = swap;
+
+ swap = d.y.vals[index];
+ d.y.vals[index] = d.y.vals[i];
+ d.y.vals[i] = swap;
+ }
+}
+
+void normalize_data_rows(data d)
+{
+ int i;
+ for(i = 0; i < d.X.rows; ++i){
+ normalize_array(d.X.vals[i], d.X.cols);
+ }
+}
+
+data *cv_split_data(data d, int part, int total)
+{
+ data *split = calloc(2, sizeof(data));
+ int i;
+ int start = part*d.X.rows/total;
+ int end = (part+1)*d.X.rows/total;
+ data train;
+ data test;
+ train.shallow = test.shallow = 1;
+
+ test.X.rows = test.y.rows = end-start;
+ train.X.rows = train.y.rows = d.X.rows - (end-start);
+ train.X.cols = test.X.cols = d.X.cols;
+ train.y.cols = test.y.cols = d.y.cols;
+ for(i = 0; i < start; ++i){
+ train.X.vals[i] = d.X.vals[i];
+ train.y.vals[i] = d.y.vals[i];
+ }
+ for(i = start; i < end; ++i){
+ test.X.vals[i-start] = d.X.vals[i];
+ test.y.vals[i-start] = d.y.vals[i];
+ }
+ for(i = end; i < d.X.rows; ++i){
+ train.X.vals[i-(start-end)] = d.X.vals[i];
+ train.y.vals[i-(start-end)] = d.y.vals[i];
+ }
+ split[0] = train;
+ split[1] = test;
+ return split;
+}
+
diff --git a/src/data.h b/src/data.h
index c01384c..3c16574 100644
--- a/src/data.h
+++ b/src/data.h
@@ -1,18 +1,24 @@
#ifndef DATA_H
#define DATA_H
-#include "image.h"
+#include "matrix.h"
typedef struct{
- int n;
- image *images;
- double **truth;
-} batch;
+ matrix X;
+ matrix y;
+ int shallow;
+} data;
-batch get_all_data(char *filename, char **labels, int k);
-batch random_batch(char *filename, int n, char **labels, int k);
-batch get_batch(char *filename, int curr, int total, char **labels, int k);
-void free_batch(batch b);
+data load_data_image_pathfile(char *filename, char **labels, int k);
+void free_data(data d);
+data load_data_image_pathfile(char *filename, char **labels, int k);
+data load_data_image_pathfile_part(char *filename, int part, int total,
+ char **labels, int k);
+data load_data_image_pathfile_random(char *filename, int n, char **labels, int k);
+data load_categorical_data_csv(char *filename, int target, int k);
+void normalize_data_rows(data d);
+void randomize_data(data d);
+data *cv_split_data(data d, int part, int total);
#endif
diff --git a/src/matrix.c b/src/matrix.c
index 562a364..5627b87 100644
--- a/src/matrix.c
+++ b/src/matrix.c
@@ -13,6 +13,17 @@
free(m.vals);
}
+void matrix_add_matrix(matrix from, matrix to)
+{
+ assert(from.rows == to.rows && from.cols == to.cols);
+ int i,j;
+ for(i = 0; i < from.rows; ++i){
+ for(j = 0; j < from.cols; ++j){
+ to.vals[i][j] += from.vals[i][j];
+ }
+ }
+}
+
matrix make_matrix(int rows, int cols)
{
matrix m;
diff --git a/src/network.c b/src/network.c
index faedb8c..29234da 100644
--- a/src/network.c
+++ b/src/network.c
@@ -15,6 +15,8 @@
net.n = n;
net.layers = calloc(net.n, sizeof(void *));
net.types = calloc(net.n, sizeof(LAYER_TYPE));
+ net.outputs = 0;
+ net.output = 0;
return net;
}
@@ -45,13 +47,13 @@
}
}
-void update_network(network net, double step)
+void update_network(network net, double step, double momentum, double decay)
{
int i;
for(i = 0; i < net.n; ++i){
if(net.types[i] == CONVOLUTIONAL){
convolutional_layer layer = *(convolutional_layer *)net.layers[i];
- update_convolutional_layer(layer, step, 0.9, .01);
+ update_convolutional_layer(layer, step, momentum, decay);
}
else if(net.types[i] == MAXPOOL){
//maxpool_layer layer = *(maxpool_layer *)net.layers[i];
@@ -61,7 +63,7 @@
}
else if(net.types[i] == CONNECTED){
connected_layer layer = *(connected_layer *)net.layers[i];
- update_connected_layer(layer, step, .9, 0);
+ update_connected_layer(layer, step, momentum, decay);
}
}
}
@@ -111,8 +113,26 @@
return get_network_delta_layer(net, net.n-1);
}
-void learn_network(network net, double *input)
+void calculate_error_network(network net, double *truth)
{
+ double *delta = get_network_delta(net);
+ double *out = get_network_output(net);
+ int i, k = get_network_output_size(net);
+ for(i = 0; i < k; ++i){
+ delta[i] = truth[i] - out[i];
+ }
+}
+
+int get_predicted_class_network(network net)
+{
+ double *out = get_network_output(net);
+ int k = get_network_output_size(net);
+ return max_index(out, k);
+}
+
+void backward_network(network net, double *input, double *truth)
+{
+ calculate_error_network(net, truth);
int i;
double *prev_input;
double *prev_delta;
@@ -145,40 +165,43 @@
}
}
-void train_network_batch(network net, batch b)
+int train_network_datum(network net, double *x, double *y, double step, double momentum, double decay)
{
- int i,j;
- int k = get_network_output_size(net);
+ forward_network(net, x);
+ int class = get_predicted_class_network(net);
+ backward_network(net, x, y);
+ update_network(net, step, momentum, decay);
+ return (y[class]?1:0);
+}
+
+double train_network_sgd(network net, data d, double step, double momentum,double decay)
+{
+ int i;
int correct = 0;
- for(i = 0; i < b.n; ++i){
- show_image(b.images[i], "Input");
- forward_network(net, b.images[i].data);
- image o = get_network_image(net);
- if(o.h) show_image_collapsed(o, "Output");
- double *output = get_network_output(net);
- double *delta = get_network_delta(net);
- int max_k = 0;
- double max = 0;
- for(j = 0; j < k; ++j){
- delta[j] = b.truth[i][j]-output[j];
- if(output[j] > max) {
- max = output[j];
- max_k = j;
- }
+ for(i = 0; i < d.X.rows; ++i){
+ int index = rand()%d.X.rows;
+ correct += train_network_datum(net, d.X.vals[index], d.y.vals[index], step, momentum, decay);
+ if((i+1)%10 == 0){
+ printf("%d: %f\n", (i+1), (double)correct/(i+1));
}
- if(b.truth[i][max_k]) ++correct;
- printf("%f\n", (double)correct/(i+1));
- learn_network(net, b.images[i].data);
- update_network(net, .001);
+ }
+ return (double)correct/d.X.rows;
+}
+
+void train_network(network net, data d, double step, double momentum, double decay)
+{
+ int i;
+ int correct = 0;
+ for(i = 0; i < d.X.rows; ++i){
+ correct += train_network_datum(net, d.X.vals[i], d.y.vals[i], step, momentum, decay);
if(i%100 == 0){
visualize_network(net);
- cvWaitKey(100);
+ cvWaitKey(10);
}
}
visualize_network(net);
- print_network(net);
cvWaitKey(100);
- printf("Accuracy: %f\n", (double)correct/b.n);
+ printf("Accuracy: %f\n", (double)correct/d.X.rows);
}
int get_network_output_size_layer(network net, int i)
@@ -250,7 +273,7 @@
{
int i,j;
for(i = 0; i < net.n; ++i){
- double *output;
+ double *output = 0;
int n = 0;
if(net.types[i] == CONVOLUTIONAL){
convolutional_layer layer = *(convolutional_layer *)net.layers[i];
@@ -283,3 +306,17 @@
fprintf(stderr, "\n");
}
}
+double network_accuracy(network net, data d)
+{
+ int i;
+ int correct = 0;
+ int k = get_network_output_size(net);
+ for(i = 0; i < d.X.rows; ++i){
+ forward_network(net, d.X.vals[i]);
+ double *out = get_network_output(net);
+ int guess = max_index(out, k);
+ if(d.y.vals[i][guess]) ++correct;
+ }
+ return (double)correct/d.X.rows;
+}
+
diff --git a/src/network.h b/src/network.h
index c655c91..3614c52 100644
--- a/src/network.h
+++ b/src/network.h
@@ -16,13 +16,17 @@
int n;
void **layers;
LAYER_TYPE *types;
+ int outputs;
+ double *output;
} network;
network make_network(int n);
void forward_network(network net, double *input);
-void learn_network(network net, double *input);
-void update_network(network net, double step);
-void train_network_batch(network net, batch b);
+void backward_network(network net, double *input, double *truth);
+void update_network(network net, double step, double momentum, double decay);
+double train_network_sgd(network net, data d, double step, double momentum,double decay);
+void train_network(network net, data d, double step, double momentum, double decay);
+double network_accuracy(network net, data d);
double *get_network_output(network net);
double *get_network_output_layer(network net, int i);
double *get_network_delta_layer(network net, int i);
@@ -31,6 +35,7 @@
int get_network_output_size(network net);
image get_network_image(network net);
image get_network_image_layer(network net, int i);
+int get_predicted_class_network(network net);
void print_network(network net);
void visualize_network(network net);
diff --git a/src/parser.c b/src/parser.c
index dc1db2b..eeb6f93 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -107,6 +107,8 @@
++count;
n = n->next;
}
+ net.outputs = get_network_output_size(net);
+ net.output = get_network_output(net);
return net;
}
diff --git a/src/tests.c b/src/tests.c
index c221042..d7d9389 100644
--- a/src/tests.c
+++ b/src/tests.c
@@ -166,19 +166,16 @@
avgerr = .99 * avgerr + .01 * err;
if(count % 1000000 == 0) printf("%f %f :%f AVG %f \n", truth, out[0], err, avgerr);
delta[0] = truth - out[0];
- learn_network(net, input);
- update_network(net, .001);
+ backward_network(net, input, &truth);
+ update_network(net, .001,0,0);
}
}
void test_data()
{
char *labels[] = {"cat","dog"};
- batch train = random_batch("train_paths.txt", 101,labels, 2);
- show_image(train.images[0], "Test Data Loading");
- show_image(train.images[100], "Test Data Loading");
- show_image(train.images[10], "Test Data Loading");
- free_batch(train);
+ data train = load_data_image_pathfile_random("train_paths.txt", 101,labels, 2);
+ free_data(train);
}
void test_full()
@@ -188,110 +185,37 @@
int i = 0;
char *labels[] = {"cat","dog"};
while(i++ < 1000 || 1){
- batch train = random_batch("train_paths.txt", 1000, labels, 2);
- train_network_batch(net, train);
- free_batch(train);
+ data train = load_data_image_pathfile_random("train_paths.txt", 1000, labels, 2);
+ train_network(net, train, .0005, 0, 0);
+ free_data(train);
printf("Round %d\n", i);
}
}
-double error_network(network net, matrix m, double **truth)
-{
- int i;
- int correct = 0;
- int k = get_network_output_size(net);
- for(i = 0; i < m.rows; ++i){
- forward_network(net, m.vals[i]);
- double *out = get_network_output(net);
- int guess = max_index(out, k);
- if(truth[i][guess]) ++correct;
- }
- return (double)correct/m.rows;
-}
-
-double **one_hot(double *a, int n, int k)
-{
- int i;
- double **t = calloc(n, sizeof(double*));
- for(i = 0; i < n; ++i){
- t[i] = calloc(k, sizeof(double));
- int index = (int)a[i];
- t[i][index] = 1;
- }
- return t;
-}
-
void test_nist()
{
- srand(999999);
+ srand(444444);
network net = parse_network_cfg("nist.cfg");
- matrix m = csv_to_matrix("mnist/mnist_train.csv");
- matrix test = csv_to_matrix("mnist/mnist_test.csv");
- double *truth_1d = pop_column(&m, 0);
- double **truth = one_hot(truth_1d, m.rows, 10);
- double *test_truth_1d = pop_column(&test, 0);
- double **test_truth = one_hot(test_truth_1d, test.rows, 10);
- int i,j;
- clock_t start = clock(), end;
- for(i = 0; i < test.rows; ++i){
- normalize_array(test.vals[i], 28*28);
- //scale_array(m.vals[i], 28*28, 1./255.);
- //translate_array(m.vals[i], 28*28, -.1);
- }
- for(i = 0; i < m.rows; ++i){
- normalize_array(m.vals[i], 28*28);
- //scale_array(m.vals[i], 28*28, 1./255.);
- //translate_array(m.vals[i], 28*28, -.1);
- }
+ data train = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
+ data test = load_categorical_data_csv("mnist/mnist_test.csv",0,10);
+ normalize_data_rows(train);
+ normalize_data_rows(test);
+ randomize_data(train);
int count = 0;
double lr = .0005;
- while(++count <= 300){
- //lr *= .99;
- int index = 0;
- int correct = 0;
- int number = 1000;
- for(i = 0; i < number; ++i){
- index = rand()%m.rows;
- forward_network(net, m.vals[index]);
- double *out = get_network_output(net);
- double *delta = get_network_delta(net);
- int max_i = 0;
- double max = out[0];
- for(j = 0; j < 10; ++j){
- delta[j] = truth[index][j]-out[j];
- if(out[j] > max){
- max = out[j];
- max_i = j;
- }
- }
- if(truth[index][max_i]) ++correct;
- learn_network(net, m.vals[index]);
- update_network(net, lr);
- }
- print_network(net);
- image input = double_to_image(28,28,1, m.vals[index]);
- //show_image(input, "Input");
- image o = get_network_image(net);
- //show_image_collapsed(o, "Output");
- visualize_network(net);
- cvWaitKey(10);
- //double test_acc = error_network(net, m, truth);
- fprintf(stderr, "\n%5d: %f %f\n\n",count, (double)correct/number, lr);
- if(count % 10 == 0 && 0){
- double train_acc = error_network(net, m, truth);
- fprintf(stderr, "\nTRAIN: %f\n", train_acc);
- double test_acc = error_network(net, test, test_truth);
- fprintf(stderr, "TEST: %f\n\n", test_acc);
- printf("%d, %f, %f\n", count, train_acc, test_acc);
- }
- if(count % (m.rows/number) == 0) lr /= 2;
+ while(++count <= 1){
+ double acc = train_network_sgd(net, train, lr, .9, .001);
+ printf("Training Accuracy: %lf", acc);
+ lr /= 2;
}
- double train_acc = error_network(net, m, truth);
- fprintf(stderr, "\nTRAIN: %f\n", train_acc);
- double test_acc = error_network(net, test, test_truth);
- fprintf(stderr, "TEST: %f\n\n", test_acc);
- printf("%d, %f, %f\n", count, train_acc, test_acc);
- end = clock();
+ /*
+ double train_acc = network_accuracy(net, train);
+ fprintf(stderr, "\nTRAIN: %f\n", train_acc);
+ double test_acc = network_accuracy(net, test);
+ fprintf(stderr, "TEST: %f\n\n", test_acc);
+ printf("%d, %f, %f\n", count, train_acc, test_acc);
+ */
+ //end = clock();
//printf("Neural Net Learning: %lf seconds\n", (double)(end-start)/CLOCKS_PER_SEC);
}
@@ -315,9 +239,9 @@
{
network net = parse_network_cfg("connected.cfg");
matrix m = csv_to_matrix("train.csv");
- matrix ho = hold_out_matrix(&m, 2500);
+ //matrix ho = hold_out_matrix(&m, 2500);
double *truth = pop_column(&m, 0);
- double *ho_truth = pop_column(&ho, 0);
+ //double *ho_truth = pop_column(&ho, 0);
int i;
clock_t start = clock(), end;
int count = 0;
@@ -333,8 +257,8 @@
delta[0] = truth[index] - out[0];
// printf("%f\n", delta[0]);
//printf("%f %f\n", truth[index], out[0]);
- learn_network(net, m.vals[index]);
- update_network(net, .00001);
+ //backward_network(net, m.vals[index], );
+ update_network(net, .00001, 0,0);
}
//double test_acc = error_network(net, m, truth);
//double valid_acc = error_network(net, ho, ho_truth);
@@ -356,33 +280,19 @@
printf("Neural Net Learning: %lf seconds\n", (double)(end-start)/CLOCKS_PER_SEC);
}
-void test_random_preprocess()
+void test_split()
{
- FILE *file = fopen("train.csv", "w");
- char *labels[] = {"cat","dog"};
- int i,j,k;
- srand(0);
- network net = parse_network_cfg("convolutional.cfg");
- for(i = 0; i < 100; ++i){
- printf("%d\n", i);
- batch part = get_batch("train_paths.txt", i, 100, labels, 2);
- for(j = 0; j < part.n; ++j){
- forward_network(net, part.images[j].data);
- double *out = get_network_output(net);
- fprintf(file, "%f", part.truth[j][0]);
- for(k = 0; k < get_network_output_size(net); ++k){
- fprintf(file, ",%f", out[k]);
- }
- fprintf(file, "\n");
- }
- free_batch(part);
- }
+ data train = load_categorical_data_csv("mnist/mnist_train.csv", 0, 10);
+ data *split = cv_split_data(train, 0, 13);
+ printf("%d, %d, %d\n", train.X.rows, split[0].X.rows, split[1].X.rows);
}
+
int main()
{
//test_kernel_update();
- test_nist();
+ test_split();
+ // test_nist();
//test_full();
//test_random_preprocess();
//test_random_classify();
diff --git a/src/utils.c b/src/utils.c
index 3b8b5a8..5180fe6 100644
--- a/src/utils.c
+++ b/src/utils.c
@@ -216,3 +216,16 @@
for(i = 0; i < 12; ++i) sum += (double)rand()/RAND_MAX;
return sum-6.;
}
+
+double **one_hot_encode(double *a, int n, int k)
+{
+ int i;
+ double **t = calloc(n, sizeof(double*));
+ for(i = 0; i < n; ++i){
+ t[i] = calloc(k, sizeof(double));
+ int index = (int)a[i];
+ t[i][index] = 1;
+ }
+ return t;
+}
+
diff --git a/src/utils.h b/src/utils.h
index 04747a4..cf38016 100644
--- a/src/utils.h
+++ b/src/utils.h
@@ -22,5 +22,6 @@
double rand_normal();
double mean_array(double *a, int n);
double variance_array(double *a, int n);
+double **one_hot_encode(double *a, int n, int k);
#endif
--
Gitblit v1.10.0