augustin64/tipe
1
1
Fork 0
mirror of https://github.com/augustin64/projet-tipe synced 2025-02-02 19:39:39 +01:00
Code Issues Packages Projects Releases Wiki Activity

Change 'input_units' to 'size_input'

Browse Source
This commit is contained in:
julienChemillier 2023-02-19 12:50:27 +01:00
parent 3d1a5a5021
commit c67d2bf697
7 changed files with 43 additions and 43 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
doc/cnn/neuron_io.md
View File

@ -44,7 +44,7 @@ type | nom de la variable | commentaire
:---:|:---:|:---: :---:|:---:|:---:
uint32_t|activation| uint32_t|activation|
uint32_t|linearisation| uint32_t|linearisation|
uint32_t|input_units| uint32_t|size_input|
uint32_t|output_units| uint32_t|output_units|
#### Si la couche est de type pooling: #### Si la couche est de type pooling:
@ -79,4 +79,4 @@ float|...|
float|bias[nn->output_units-1]|biais float|bias[nn->output_units-1]|biais
float|weights[0][0]|poids float|weights[0][0]|poids
float|...| float|...|
float|weights[nn->input_units-1][nn->output_units-1]| float|weights[nn->size_input-1][nn->output_units-1]|

30
src/cnn/creation.c
View File

@ -205,7 +205,7 @@ void add_convolution(Network* network, int depth_output, int dim_output, int act
void add_dense(Network* network, int output_units, int activation) { void add_dense(Network* network, int output_units, int activation) {
int n = network->size; int n = network->size;
int k_pos = n-1; int k_pos = n-1;
int input_units = network->width[k_pos]; int size_input = network->width[k_pos];
if (network->max_size == n) { if (network->max_size == n) {
printf("Impossible de rajouter une couche dense, le réseau est déjà plein\n"); printf("Impossible de rajouter une couche dense, le réseau est déjà plein\n");
return; return;
@ -216,7 +216,7 @@ void add_dense(Network* network, int output_units, int activation) {
network->kernel[k_pos]->activation = activation; network->kernel[k_pos]->activation = activation;
network->kernel[k_pos]->linearisation = 0; network->kernel[k_pos]->linearisation = 0;
network->kernel[k_pos]->pooling = 0; network->kernel[k_pos]->pooling = 0;
nn->input_units = input_units; nn->size_input = size_input;
nn->output_units = output_units; nn->output_units = output_units;
nn->bias = (float*)nalloc(sizeof(float)*output_units); nn->bias = (float*)nalloc(sizeof(float)*output_units);
nn->d_bias = (float*)nalloc(sizeof(float)*output_units); nn->d_bias = (float*)nalloc(sizeof(float)*output_units);
@ -224,9 +224,9 @@ void add_dense(Network* network, int output_units, int activation) {
nn->d_bias[i] = 0.; nn->d_bias[i] = 0.;
} }
nn->weights = (float**)nalloc(sizeof(float*)*input_units); nn->weights = (float**)nalloc(sizeof(float*)*size_input);
nn->d_weights = (float**)nalloc(sizeof(float*)*input_units); nn->d_weights = (float**)nalloc(sizeof(float*)*size_input);
for (int i=0; i < input_units; i++) { for (int i=0; i < size_input; i++) {
nn->weights[i] = (float*)nalloc(sizeof(float)*output_units); nn->weights[i] = (float*)nalloc(sizeof(float)*output_units);
nn->d_weights[i] = (float*)nalloc(sizeof(float)*output_units); nn->d_weights[i] = (float*)nalloc(sizeof(float)*output_units);
for (int j=0; j < output_units; j++) { for (int j=0; j < output_units; j++) {
@ -234,19 +234,19 @@ void add_dense(Network* network, int output_units, int activation) {
} }
} }
initialisation_1d_matrix(network->initialisation, nn->bias, output_units, input_units); initialisation_1d_matrix(network->initialisation, nn->bias, output_units, size_input);
initialisation_2d_matrix(network->initialisation, nn->weights, input_units, output_units, input_units, output_units); initialisation_2d_matrix(network->initialisation, nn->weights, size_input, output_units, size_input, output_units);
create_a_line_input_layer(network, n, output_units); create_a_line_input_layer(network, n, output_units);
create_a_line_input_z_layer(network, n, output_units); create_a_line_input_z_layer(network, n, output_units);
network->size++; network->size++;
} }
void add_dense_linearisation(Network* network, int output_units, int activation) { void add_dense_linearisation(Network* network, int output_units, int activation) {
// Can replace input_units by a research of this dim // Can replace size_input by a research of this dim
int n = network->size; int n = network->size;
int k_pos = n-1; int k_pos = n-1;
int input_units = network->depth[k_pos]*network->width[k_pos]*network->width[k_pos]; int size_input = network->depth[k_pos]*network->width[k_pos]*network->width[k_pos];
if (network->max_size == n) { if (network->max_size == n) {
printf("Impossible de rajouter une couche dense, le réseau est déjà plein\n"); printf("Impossible de rajouter une couche dense, le réseau est déjà plein\n");
return; return;
@ -257,7 +257,7 @@ void add_dense_linearisation(Network* network, int output_units, int activation)
network->kernel[k_pos]->activation = activation; network->kernel[k_pos]->activation = activation;
network->kernel[k_pos]->linearisation = 1; network->kernel[k_pos]->linearisation = 1;
network->kernel[k_pos]->pooling = 0; network->kernel[k_pos]->pooling = 0;
nn->input_units = input_units; nn->size_input = size_input;
nn->output_units = output_units; nn->output_units = output_units;
nn->bias = (float*)nalloc(sizeof(float)*output_units); nn->bias = (float*)nalloc(sizeof(float)*output_units);
@ -265,17 +265,17 @@ void add_dense_linearisation(Network* network, int output_units, int activation)
for (int i=0; i < output_units; i++) { for (int i=0; i < output_units; i++) {
nn->d_bias[i] = 0.; nn->d_bias[i] = 0.;
} }
nn->weights = (float**)nalloc(sizeof(float*)*input_units); nn->weights = (float**)nalloc(sizeof(float*)*size_input);
nn->d_weights = (float**)nalloc(sizeof(float*)*input_units); nn->d_weights = (float**)nalloc(sizeof(float*)*size_input);
for (int i=0; i < input_units; i++) { for (int i=0; i < size_input; i++) {
nn->weights[i] = (float*)nalloc(sizeof(float)*output_units); nn->weights[i] = (float*)nalloc(sizeof(float)*output_units);
nn->d_weights[i] = (float*)nalloc(sizeof(float)*output_units); nn->d_weights[i] = (float*)nalloc(sizeof(float)*output_units);
for (int j=0; j < output_units; j++) { for (int j=0; j < output_units; j++) {
nn->d_weights[i][j] = 0.; nn->d_weights[i][j] = 0.;
} }
} }
initialisation_1d_matrix(network->initialisation, nn->bias, output_units, input_units); initialisation_1d_matrix(network->initialisation, nn->bias, output_units, size_input);
initialisation_2d_matrix(network->initialisation, nn->weights, input_units, output_units, input_units, output_units); initialisation_2d_matrix(network->initialisation, nn->weights, size_input, output_units, size_input, output_units);
create_a_line_input_layer(network, n, output_units); create_a_line_input_layer(network, n, output_units);
create_a_line_input_z_layer(network, n, output_units); create_a_line_input_z_layer(network, n, output_units);
network->size++; network->size++;

4
src/cnn/free.c
View File

@ -70,7 +70,7 @@ void free_convolution(Network* network, int pos) {
void free_dense(Network* network, int pos) { void free_dense(Network* network, int pos) {
free_a_line_input_layer(network, pos+1); free_a_line_input_layer(network, pos+1);
Kernel_nn* k_pos = network->kernel[pos]->nn; Kernel_nn* k_pos = network->kernel[pos]->nn;
int dim = k_pos->input_units; int dim = k_pos->size_input;
for (int i=0; i < dim; i++) { for (int i=0; i < dim; i++) {
gree(k_pos->weights[i]); gree(k_pos->weights[i]);
gree(k_pos->d_weights[i]); gree(k_pos->d_weights[i]);
@ -87,7 +87,7 @@ void free_dense(Network* network, int pos) {
void free_dense_linearisation(Network* network, int pos) { void free_dense_linearisation(Network* network, int pos) {
free_a_line_input_layer(network, pos+1); free_a_line_input_layer(network, pos+1);
Kernel_nn* k_pos = network->kernel[pos]->nn; Kernel_nn* k_pos = network->kernel[pos]->nn;
int dim = k_pos->input_units; int dim = k_pos->size_input;
for (int i=0; i < dim; i++) { for (int i=0; i < dim; i++) {
gree(k_pos->weights[i]); gree(k_pos->weights[i]);
gree(k_pos->d_weights[i]); gree(k_pos->d_weights[i]);

6
src/cnn/include/struct.h
View File

@ -12,12 +12,12 @@ typedef struct Kernel_cnn {
} Kernel_cnn; } Kernel_cnn;
typedef struct Kernel_nn { typedef struct Kernel_nn {
int input_units; // Nombre d'éléments en entrée int size_input; // Nombre d'éléments en entrée
int output_units; // Nombre d'éléments en sortie int output_units; // Nombre d'éléments en sortie
float* bias; // bias[output_units] float* bias; // bias[output_units]
float* d_bias; // d_bias[output_units] float* d_bias; // d_bias[output_units]
float** weights; // weight[input_units][output_units] float** weights; // weight[size_input][output_units]
float** d_weights; // d_weights[input_units][output_units] float** d_weights; // d_weights[size_input][output_units]
} Kernel_nn; } Kernel_nn;
typedef struct Kernel { typedef struct Kernel {

12
src/cnn/neuron_io.c
View File

@ -111,7 +111,7 @@ void write_couche(Network* network, int indice_couche, int type_couche, FILE* pt
uint32_t pre_buffer[4]; uint32_t pre_buffer[4];
pre_buffer[0] = kernel->activation; pre_buffer[0] = kernel->activation;
pre_buffer[1] = kernel->linearisation; pre_buffer[1] = kernel->linearisation;
pre_buffer[2] = nn->input_units; pre_buffer[2] = nn->size_input;
pre_buffer[3] = nn->output_units; pre_buffer[3] = nn->output_units;
fwrite(pre_buffer, sizeof(pre_buffer), 1, ptr); fwrite(pre_buffer, sizeof(pre_buffer), 1, ptr);
@ -122,7 +122,7 @@ void write_couche(Network* network, int indice_couche, int type_couche, FILE* pt
} }
fwrite(buffer, sizeof(buffer), 1, ptr); fwrite(buffer, sizeof(buffer), 1, ptr);
for (int i=0; i < nn->input_units; i++) { for (int i=0; i < nn->size_input; i++) {
indice_buffer = 0; indice_buffer = 0;
float buffer[nn->output_units]; float buffer[nn->output_units];
for (int j=0; j < nn->output_units; j++) { for (int j=0; j < nn->output_units; j++) {
@ -287,7 +287,7 @@ Kernel* read_kernel(int type_couche, int output_dim, FILE* ptr) {
kernel->activation = buffer[0]; kernel->activation = buffer[0];
kernel->linearisation = buffer[1]; kernel->linearisation = buffer[1];
kernel->nn->input_units = buffer[2]; kernel->nn->size_input = buffer[2];
kernel->nn->output_units = buffer[3]; kernel->nn->output_units = buffer[3];
// Lecture du corps // Lecture du corps
@ -302,9 +302,9 @@ Kernel* read_kernel(int type_couche, int output_dim, FILE* ptr) {
nn->d_bias[i] = 0.; nn->d_bias[i] = 0.;
} }
nn->weights = (float**)nalloc(sizeof(float*)*nn->input_units); nn->weights = (float**)nalloc(sizeof(float*)*nn->size_input);
nn->d_weights = (float**)nalloc(sizeof(float*)*nn->input_units); nn->d_weights = (float**)nalloc(sizeof(float*)*nn->size_input);
for (int i=0; i < nn->input_units; i++) { for (int i=0; i < nn->size_input; i++) {
nn->weights[i] = (float*)nalloc(sizeof(float)*nn->output_units); nn->weights[i] = (float*)nalloc(sizeof(float)*nn->output_units);
nn->d_weights[i] = (float*)nalloc(sizeof(float)*nn->output_units); nn->d_weights[i] = (float*)nalloc(sizeof(float)*nn->output_units);
for (int j=0; j < nn->output_units; j++) { for (int j=0; j < nn->output_units; j++) {

28
src/cnn/utils.c
View File

@ -56,12 +56,12 @@ bool equals_networks(Network* network1, Network* network2) {
checkEquals(kernel[i]->pooling, "kernel[i]->pooling pour un pooling", i); checkEquals(kernel[i]->pooling, "kernel[i]->pooling pour un pooling", i);
} else if (!network1->kernel[i]->cnn) { } else if (!network1->kernel[i]->cnn) {
// Type NN // Type NN
checkEquals(kernel[i]->nn->input_units, "kernel[i]->nn->input_units", i); checkEquals(kernel[i]->nn->size_input, "kernel[i]->nn->size_input", i);
checkEquals(kernel[i]->nn->output_units, "kernel[i]->nn->output_units", i); checkEquals(kernel[i]->nn->output_units, "kernel[i]->nn->output_units", i);
for (int j=0; j < network1->kernel[i]->nn->output_units; j++) { for (int j=0; j < network1->kernel[i]->nn->output_units; j++) {
checkEquals(kernel[i]->nn->bias[j], "kernel[i]->nn->bias[j]", j); checkEquals(kernel[i]->nn->bias[j], "kernel[i]->nn->bias[j]", j);
} }
for (int j=0; j < network1->kernel[i]->nn->input_units; j++) { for (int j=0; j < network1->kernel[i]->nn->size_input; j++) {
for (int k=0; k < network1->kernel[i]->nn->output_units; k++) { for (int k=0; k < network1->kernel[i]->nn->output_units; k++) {
checkEquals(kernel[i]->nn->weights[j][k], "kernel[i]->nn->weights[j][k]", k); checkEquals(kernel[i]->nn->weights[j][k], "kernel[i]->nn->weights[j][k]", k);
} }
@ -100,7 +100,7 @@ Network* copy_network(Network* network) {
// Paramètre du réseau // Paramètre du réseau
int size = network->size; int size = network->size;
// Paramètres des couches NN // Paramètres des couches NN
int input_units; int size_input;
int output_units; int output_units;
// Paramètres des couches CNN // Paramètres des couches CNN
int rows; int rows;
@ -137,13 +137,13 @@ Network* copy_network(Network* network) {
copyVar(kernel[i]->activation); copyVar(kernel[i]->activation);
copyVar(kernel[i]->linearisation); // 0 copyVar(kernel[i]->linearisation); // 0
input_units = network->kernel[i]->nn->input_units; size_input = network->kernel[i]->nn->size_input;
output_units = network->kernel[i]->nn->output_units; output_units = network->kernel[i]->nn->output_units;
network_cp->kernel[i]->cnn = NULL; network_cp->kernel[i]->cnn = NULL;
network_cp->kernel[i]->nn = (Kernel_nn*)nalloc(sizeof(Kernel_nn)); network_cp->kernel[i]->nn = (Kernel_nn*)nalloc(sizeof(Kernel_nn));
copyVar(kernel[i]->nn->input_units); copyVar(kernel[i]->nn->size_input);
copyVar(kernel[i]->nn->output_units); copyVar(kernel[i]->nn->output_units);
network_cp->kernel[i]->nn->bias = (float*)nalloc(sizeof(float)*output_units); network_cp->kernel[i]->nn->bias = (float*)nalloc(sizeof(float)*output_units);
@ -153,9 +153,9 @@ Network* copy_network(Network* network) {
network_cp->kernel[i]->nn->d_bias[j] = 0.; network_cp->kernel[i]->nn->d_bias[j] = 0.;
} }
network_cp->kernel[i]->nn->weights = (float**)nalloc(sizeof(float*)*input_units); network_cp->kernel[i]->nn->weights = (float**)nalloc(sizeof(float*)*size_input);
network_cp->kernel[i]->nn->d_weights = (float**)nalloc(sizeof(float*)*input_units); network_cp->kernel[i]->nn->d_weights = (float**)nalloc(sizeof(float*)*size_input);
for (int j=0; j < input_units; j++) { for (int j=0; j < size_input; j++) {
network_cp->kernel[i]->nn->weights[j] = (float*)nalloc(sizeof(float)*output_units); network_cp->kernel[i]->nn->weights[j] = (float*)nalloc(sizeof(float)*output_units);
network_cp->kernel[i]->nn->d_weights[j] = (float*)nalloc(sizeof(float)*output_units); network_cp->kernel[i]->nn->d_weights[j] = (float*)nalloc(sizeof(float)*output_units);
for (int k=0; k < output_units; k++) { for (int k=0; k < output_units; k++) {
@ -254,7 +254,7 @@ void copy_network_parameters(Network* network_src, Network* network_dest) {
// Paramètre du réseau // Paramètre du réseau
int size = network_src->size; int size = network_src->size;
// Paramètres des couches NN // Paramètres des couches NN
int input_units; int size_input;
int output_units; int output_units;
// Paramètres des couches CNN // Paramètres des couches CNN
int rows; int rows;
@ -267,13 +267,13 @@ void copy_network_parameters(Network* network_src, Network* network_dest) {
for (int i=0; i < size-1; i++) { for (int i=0; i < size-1; i++) {
if (!network_src->kernel[i]->cnn && network_src->kernel[i]->nn) { // Cas du NN if (!network_src->kernel[i]->cnn && network_src->kernel[i]->nn) { // Cas du NN
input_units = network_src->kernel[i]->nn->input_units; size_input = network_src->kernel[i]->nn->size_input;
output_units = network_src->kernel[i]->nn->output_units; output_units = network_src->kernel[i]->nn->output_units;
for (int j=0; j < output_units; j++) { for (int j=0; j < output_units; j++) {
copyVarParams(kernel[i]->nn->bias[j]); copyVarParams(kernel[i]->nn->bias[j]);
} }
for (int j=0; j < input_units; j++) { for (int j=0; j < size_input; j++) {
for (int k=0; k < output_units; k++) { for (int k=0; k < output_units; k++) {
copyVarParams(kernel[i]->nn->weights[j][k]); copyVarParams(kernel[i]->nn->weights[j][k]);
} }
@ -315,7 +315,7 @@ int count_null_weights(Network* network) {
int size = network->size; int size = network->size;
// Paramètres des couches NN // Paramètres des couches NN
int input_units; int size_input;
int output_units; int output_units;
// Paramètres des couches CNN // Paramètres des couches CNN
int rows; int rows;
@ -326,13 +326,13 @@ int count_null_weights(Network* network) {
for (int i=0; i < size-1; i++) { for (int i=0; i < size-1; i++) {
if (!network->kernel[i]->cnn && network->kernel[i]->nn) { // Cas du NN if (!network->kernel[i]->cnn && network->kernel[i]->nn) { // Cas du NN
input_units = network->kernel[i]->nn->input_units; size_input = network->kernel[i]->nn->size_input;
output_units = network->kernel[i]->nn->output_units; output_units = network->kernel[i]->nn->output_units;
for (int j=0; j < output_units; j++) { for (int j=0; j < output_units; j++) {
null_bias += fabs(network->kernel[i]->nn->bias[j]) <= epsilon; null_bias += fabs(network->kernel[i]->nn->bias[j]) <= epsilon;
} }
for (int j=0; j < input_units; j++) { for (int j=0; j < size_input; j++) {
for (int k=0; k < output_units; k++) { for (int k=0; k < output_units; k++) {
null_weights += fabs(network->kernel[i]->nn->weights[j][k]) <= epsilon; null_weights += fabs(network->kernel[i]->nn->weights[j][k]) <= epsilon;
} }

2
test/cnn_structure.c
View File

@ -27,7 +27,7 @@ int main() {
} }
} else if (!kernel->cnn) { } else if (!kernel->cnn) {
printf("\n==== Couche %d de type "GREEN"NN"RESET" ====\n", i); printf("\n==== Couche %d de type "GREEN"NN"RESET" ====\n", i);
printf("input: %d\n", kernel->nn->input_units); printf("input: %d\n", kernel->nn->size_input);
printf("output: %d\n", kernel->nn->output_units); printf("output: %d\n", kernel->nn->output_units);
} else { } else {
printf("\n==== Couche %d de type "BLUE"CNN"RESET" ====\n", i); printf("\n==== Couche %d de type "BLUE"CNN"RESET" ====\n", i);