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

Revert fd36226 for cnn/creation.c

Browse Source
This commit is contained in:
augustin64 2023-05-12 16:21:14 +02:00
parent 3ac318dd2c
commit 49a2299c1c
1 changed files with 1 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
src/cnn/creation.c
View File

@ -38,7 +38,7 @@ Network* create_network(int max_size, float learning_rate, int dropout, int acti
return network; return network;
} }
/*Network* create_network_lenet5(float learning_rate, int dropout, int activation, int initialisation, int input_dim, int input_depth) { Network* create_network_lenet5(float learning_rate, int dropout, int activation, int initialisation, int input_dim, int input_depth) {
Network* network = create_network(8, learning_rate, dropout, activation, initialisation, input_dim, input_depth); Network* network = create_network(8, learning_rate, dropout, activation, initialisation, input_dim, input_depth);
add_convolution(network, 6, 28, activation); add_convolution(network, 6, 28, activation);
add_average_pooling(network, 14); add_average_pooling(network, 14);
@ -48,32 +48,8 @@ Network* create_network(int max_size, float learning_rate, int dropout, int acti
add_dense(network, 84, activation); add_dense(network, 84, activation);
add_dense(network, 10, SOFTMAX); add_dense(network, 10, SOFTMAX);
return network; return network;
}*/
Network* create_network_lenet5(float learning_rate, int dropout, int activation, int initialisation, int input_dim, int input_depth) {
input_dim = 260;
input_depth = 3;
Network* network = create_network(16, learning_rate, dropout, activation, initialisation, input_dim, input_depth);
printf_warning("Creating large network");
printf(" %d %d\n", input_dim, input_depth);
add_convolution(network, 6, 258, activation);
add_convolution(network, 16, 256, activation);
add_average_pooling(network, 64);
add_convolution(network, 16, 60, activation);
add_average_pooling(network, 30);
add_convolution(network, 16, 26, activation);
add_convolution(network, 16, 22, activation);
add_convolution(network, 16, 18, activation);
add_dense_linearisation(network, 840, activation);
add_dense(network, 520, activation);
add_dense(network, 420, activation);
add_dense(network, 320, activation);
add_dense(network, 220, activation);
add_dense(network, 120, activation);
add_dense(network, 50, SOFTMAX);
return network;
} }
Network* create_simple_one(float learning_rate, int dropout, int activation, int initialisation, int input_dim, int input_depth) { Network* create_simple_one(float learning_rate, int dropout, int activation, int initialisation, int input_dim, int input_depth) {
Network* network = create_network(3, learning_rate, dropout, activation, initialisation, input_dim, input_depth); Network* network = create_network(3, learning_rate, dropout, activation, initialisation, input_dim, input_depth);
add_dense_linearisation(network, 80, activation); add_dense_linearisation(network, 80, activation);