tipe/src/mnist/utils.c

344 lines
11 KiB
C
Raw Normal View History

2022-04-19 13:55:08 +02:00
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <float.h>
2022-09-30 15:44:28 +02:00
#include "include/neural_network.h"
#include "include/neuron_io.h"
#include "include/mnist.h"
2022-04-19 13:55:08 +02:00
2022-04-19 21:30:23 +02:00
/*
Contient un ensemble de fonctions utiles pour le débogage
*/
void help(char* call) {
2022-05-21 18:06:39 +02:00
printf("Usage: %s ( print-poids | print-biais | creer-reseau | patch-network ) [OPTIONS]\n\n", call);
2022-04-19 21:30:23 +02:00
printf("OPTIONS:\n");
printf("\tprint-poids:\n");
2022-04-30 09:31:52 +02:00
printf("\t\t--reseau | -r [FILENAME]\tFichier contenant le réseau de neurones.\n");
printf("\tprint-biais:\n");
printf("\t\t--reseau | -r [FILENAME]\tFichier contenant le réseau de neurones.\n");
2022-04-22 15:03:21 +02:00
printf("\tcount-labels:\n");
printf("\t\t--labels | -l [FILENAME]\tFichier contenant les labels.\n");
2022-04-30 09:31:52 +02:00
printf("\tcreer-reseau:\n");
printf("\t\t--out | -o [FILENAME]\tFichier où écrire le réseau de neurones.\n");
2022-05-21 18:06:39 +02:00
printf("\t\t--number | -n [int]\tNuméro à privilégier.\n");
printf("\tpatch-network:\n");
printf("\t\t--network | -n [FILENAME]\tFichier contenant le réseau de neurones.\n");
printf("\t\t--delta | -d [FILENAME]\tFichier de patch à utiliser.\n");
2022-07-01 16:16:55 +02:00
printf("\tprint-images:\n");
printf("\t\t--images | -i [FILENAME]\tFichier contenant les images.\n");
2022-07-02 10:56:06 +02:00
printf("\tprint-poids-neurone:\n");
printf("\t\t--reseau | -r [FILENAME]\tFichier contenant le réseau de neurones.\n");
printf("\t\t--neurone | -n [int]\tNuméro du neurone dont il faut afficher les poids.\n");
2022-04-19 21:30:23 +02:00
}
2022-04-25 14:39:45 +02:00
void print_bias(char* filename) {
2022-05-03 09:44:40 +02:00
Network* network = read_network(filename);
2022-04-19 13:55:08 +02:00
2022-04-25 14:39:45 +02:00
for (int i=1; i < network->nb_layers -1; i++) {
2022-04-30 09:31:52 +02:00
printf("Couche %d\n", i);
2022-04-25 14:39:45 +02:00
for (int j=0; j < network->layers[i]->nb_neurons; j++) {
2022-04-30 09:31:52 +02:00
printf("Couche %d\tNeurone %d\tBiais: %f\n", i, j, network->layers[i]->neurons[j]->bias);
2022-04-19 13:55:08 +02:00
}
}
2022-04-25 14:39:45 +02:00
deletion_of_network(network);
2022-04-19 21:30:23 +02:00
}
2022-05-10 21:04:48 +02:00
void print_weights(char* filename) {
2022-05-03 09:44:40 +02:00
Network* network = read_network(filename);
2022-04-25 14:39:45 +02:00
for (int i=0; i < network->nb_layers -1; i++) {
2022-04-30 09:31:52 +02:00
printf("Couche %d\n", i);
2022-04-25 14:39:45 +02:00
for (int j=0; j < network->layers[i]->nb_neurons; j++) {
2022-04-30 09:31:52 +02:00
printf("Couche %d\tNeurone %d\tPoids: [", i, j);
2022-04-25 14:39:45 +02:00
for (int k=0; k < network->layers[i+1]->nb_neurons; k++) {
printf("%f, ", network->layers[i]->neurons[j]->weights[k]);
2022-04-19 21:30:23 +02:00
}
printf("]\n");
}
}
2022-04-25 14:39:45 +02:00
deletion_of_network(network);
2022-04-19 13:55:08 +02:00
}
2022-04-22 15:03:21 +02:00
void count_labels(char* filename) {
uint32_t number_of_images = read_mnist_labels_nb_images(filename);
2022-05-14 14:46:48 +02:00
unsigned int* labels = (unsigned int*)malloc(sizeof(unsigned int)*number_of_images);
2022-04-22 15:03:21 +02:00
labels = read_mnist_labels(filename);
2022-05-14 14:46:48 +02:00
unsigned int tab[10];
2022-04-22 15:03:21 +02:00
for (int i=0; i < 10; i++) {
tab[i] = 0;
}
2022-05-14 14:46:48 +02:00
for (int i=0; i < (int)number_of_images; i++) {
2022-04-22 15:03:21 +02:00
tab[(int)labels[i]]++;
}
for (int i=0; i < 10; i++) {
2022-11-08 19:56:58 +01:00
printf("Nombre de %d: %u\n", i, tab[i]);
2022-04-22 15:03:21 +02:00
}
2022-05-23 17:27:38 +02:00
free(labels);
2022-04-22 15:03:21 +02:00
}
2022-04-25 14:39:45 +02:00
void create_network(char* filename, int sortie) {
2022-05-14 14:46:48 +02:00
Network* network = (Network*)malloc(sizeof(Network));
2022-04-25 14:39:45 +02:00
Layer* layer;
Neuron* neuron;
network->nb_layers = 3;
2022-04-19 21:30:23 +02:00
2022-05-14 14:46:48 +02:00
network->layers = (Layer**)malloc(sizeof(Layer*)*network->nb_layers);
2022-04-25 14:39:45 +02:00
int neurons_per_layer[4] = {784, 1, 10, 0};
for (int i=0; i < network->nb_layers; i++) {
2022-05-14 14:46:48 +02:00
layer = (Layer*)malloc(sizeof(Layer));
2022-04-25 14:39:45 +02:00
layer->nb_neurons = neurons_per_layer[i];
2022-05-14 14:46:48 +02:00
layer->neurons = (Neuron**)malloc(sizeof(Neuron*)*layer->nb_neurons);
2022-04-25 14:39:45 +02:00
for (int j=0; j < layer->nb_neurons; j++) {
2022-05-14 14:46:48 +02:00
neuron = (Neuron*)malloc(sizeof(Neuron));
2022-04-25 14:39:45 +02:00
neuron->bias = 0.;
neuron->z = 0.;
neuron->back_bias = 0.;
neuron->last_back_bias = 0.;
2022-05-14 14:46:48 +02:00
neuron->weights = (float*)malloc(sizeof(float)*neurons_per_layer[i+1]);
neuron->back_weights = (float*)malloc(sizeof(float)*neurons_per_layer[i+1]);
neuron->last_back_weights = (float*)malloc(sizeof(float)*neurons_per_layer[i+1]);
2022-04-25 14:39:45 +02:00
for (int k=0; k < neurons_per_layer[i+1]; k++) {
neuron->weights[k] = 0.;
neuron->back_weights[k] = 0.;
neuron->last_back_weights[k] = 0.;
2022-04-19 21:30:23 +02:00
}
2022-05-14 14:46:48 +02:00
layer->neurons[j] = neuron;
2022-04-19 21:30:23 +02:00
}
2022-05-14 14:46:48 +02:00
network->layers[i] = layer;
2022-04-19 21:30:23 +02:00
}
2022-04-25 14:39:45 +02:00
for (int j=0; j < neurons_per_layer[0]; j++) {
network->layers[0]->neurons[j]->weights[0] = 1;
2022-04-19 21:30:23 +02:00
}
2022-04-25 14:39:45 +02:00
network->layers[1]->neurons[0]->weights[sortie] = 1;
write_network(filename, network);
deletion_of_network(network);
2022-04-19 21:30:23 +02:00
}
2022-05-21 18:06:39 +02:00
void patch_stored_network(char* network_filename, char* delta_filename) {
// Apply patch to a network stored in a file
Network* network = read_network(network_filename);
Network* delta = read_delta_network(delta_filename);
patch_network(network, delta, 1);
write_network(network_filename, network);
deletion_of_network(network);
deletion_of_network(delta);
}
2022-04-19 21:30:23 +02:00
2022-07-01 16:16:55 +02:00
void print_images(char* filename) {
int* parameters = read_mnist_images_parameters(filename);
int nb_elem = parameters[0];
int width = parameters[1];
int height = parameters[2];
free(parameters);
int*** images = read_mnist_images(filename);
printf("[\n");
for (int i=0; i < nb_elem; i++) {
printf("\t[\n");
for (int j=0; j < height; j++) {
printf("\t\t[");
for (int k=0; k < width; k++) {
if (k != width -1)
printf("%d, ", images[i][j][k]);
else
printf("%d", images[i][j][k]);
}
if (j != height -1)
printf("],\n");
else
printf("]\n");
free(images[i][j]);
}
if (i != nb_elem -1)
printf("\t],\n");
else
printf("\t]\n");
free(images[i]);
}
free(images);
printf("]\n");
}
2022-07-02 10:56:06 +02:00
void print_poids_neurone(char* filename, int num_neurone) {
Network* network = read_network(filename);
int nb_layers = network->nb_layers;
Layer* layer = network->layers[nb_layers-2];
int nb_neurons = layer->nb_neurons;
printf("[\n");
for (int i=0; i < nb_neurons; i++) {
printf("%f", layer->neurons[i]->weights[num_neurone]);
if (i != nb_neurons -1)
printf(", ");
else
printf("\n");
}
printf("]\n");
}
2022-04-19 16:26:28 +02:00
int main(int argc, char* argv[]) {
2022-04-19 21:30:23 +02:00
if (argc < 2) {
printf("Pas d'action spécifiée\n");
help(argv[0]);
exit(1);
}
if (! strcmp(argv[1], "print-poids")) {
char* filename = NULL;
int i = 2;
while (i < argc) {
2022-04-30 09:31:52 +02:00
if ((! strcmp(argv[i], "--reseau"))||(! strcmp(argv[i], "-r"))) {
2022-04-19 21:30:23 +02:00
filename = argv[i+1];
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
if (! filename) {
2022-05-03 09:44:40 +02:00
printf("Pas de fichier spécifié, utilisation de '.cache/reseau.bin'\n");
filename = ".cache/reseau.bin";
2022-04-19 21:30:23 +02:00
}
2022-05-10 21:04:48 +02:00
print_weights(filename);
2022-05-16 17:26:04 +02:00
exit(0);
2022-04-30 09:31:52 +02:00
} else if (! strcmp(argv[1], "print-biais")) {
2022-04-19 21:30:23 +02:00
char* filename = NULL;
int i = 2;
while (i < argc) {
2022-04-30 09:31:52 +02:00
if ((! strcmp(argv[i], "--reseau"))||(! strcmp(argv[i], "-r"))) {
2022-04-19 21:30:23 +02:00
filename = argv[i+1];
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
if (! filename) {
2022-05-03 09:44:40 +02:00
printf("Pas de fichier spécifié, utilisation de '.cache/reseau.bin'\n");
filename = ".cache/reseau.bin";
2022-04-19 21:30:23 +02:00
}
2022-04-25 14:39:45 +02:00
print_bias(filename);
2022-05-16 17:26:04 +02:00
exit(0);
2022-04-30 09:31:52 +02:00
} else if (! strcmp(argv[1], "creer-reseau")) {
2022-04-19 21:30:23 +02:00
char* out = NULL;
int n = -1;
int i = 2;
while (i < argc) {
if ((! strcmp(argv[i], "--out"))||(! strcmp(argv[i], "-o"))) {
out = argv[i+1];
i += 2;
} else if ((! strcmp(argv[i], "--number"))||(! strcmp(argv[i], "-n"))) {
n = strtol(argv[i+1], NULL, 10);
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
2022-05-14 14:46:48 +02:00
create_network(out, n);
2022-05-16 17:26:04 +02:00
exit(0);
2022-04-22 15:03:21 +02:00
} else if (! strcmp(argv[1], "count-labels")) {
char* labels = NULL;
int i = 2;
while (i < argc) {
if ((! strcmp(argv[i], "--labels"))||(! strcmp(argv[i], "-l"))) {
labels = argv[i+1];
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
2022-04-19 21:30:23 +02:00
}
2022-04-22 15:03:21 +02:00
if (! labels) {
printf("Pas de fichier spécifié, défaut: 'data/mnist/train-labels-idx1-ubyte'\n");
labels = "data/mnist/train-labels-idx1-ubyte";
2022-04-19 21:30:23 +02:00
}
2022-04-22 15:03:21 +02:00
count_labels(labels);
2022-05-16 17:26:04 +02:00
exit(0);
2022-07-01 16:16:55 +02:00
} else if (! strcmp(argv[1], "patch-network")) {
2022-05-21 18:06:39 +02:00
char* network = NULL;
char* delta = NULL;
int i = 2;
while (i < argc) {
if ((! strcmp(argv[i], "--network"))||(! strcmp(argv[i], "-n"))) {
network = argv[i+1];
i += 2;
} else if ((! strcmp(argv[i], "--delta"))||(! strcmp(argv[i], "-d"))) {
delta = argv[i+1];
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
if (!network) {
printf("--network: Argument obligatoire.\n");
exit(1);
}
if (!delta) {
printf("--delta: Argument obligatoire.\n");
exit(1);
}
patch_stored_network(network, delta);
exit(0);
2022-07-01 16:16:55 +02:00
} else if (! strcmp(argv[1], "print-images")) {
char* images = NULL;
int i = 2;
while (i < argc) {
if ((! strcmp(argv[i], "--images"))||(! strcmp(argv[i], "-i"))) {
images = argv[i+1];
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
if (!images) {
printf("--images: Argument obligatoire.\n");
exit(1);
}
print_images(images);
exit(0);
2022-07-02 10:56:06 +02:00
} else if (! strcmp(argv[1], "print-poids-neurone")) {
char* reseau = NULL;
int neurone = 0;
int i = 2;
while (i < argc) {
if ((! strcmp(argv[i], "--reseau"))||(! strcmp(argv[i], "-r"))) {
reseau = argv[i+1];
i += 2;
} else if ((! strcmp(argv[i], "--neurone"))||(! strcmp(argv[i], "-n"))) {
neurone = strtol(argv[i+1], NULL, 10);
i += 2;
} else {
printf("%s : Argument non reconnu\n", argv[i]);
i++;
}
}
if (!reseau) {
printf("--reseau: Argument obligatoire.\n");
exit(1);
}
print_poids_neurone(reseau, neurone);
exit(0);
2022-04-19 21:30:23 +02:00
}
printf("Option choisie non reconnue: %s\n", argv[1]);
help(argv[0]);
2022-04-19 13:55:08 +02:00
return 1;
}