Mirrors/MLP
1
0
Fork 0
mirror of https://github.com/davidalbertonogueira/MLP.git synced 2025-12-18 12:54:41 +03:00
Code Issues Actions Packages Projects Releases Wiki Activity

Internal architecture changes (to allow diferent activation functions for each layer and to allow hidden layers to have different number of nodes).

Browse Source
This commit is contained in:
davidjacnogueira
2016-11-09 02:31:53 +00:00
parent f647b05f70
commit 0a636416ed
7 changed files with 193 additions and 100 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
src/MLPTest.cpp
View File

@@ -39,7 +39,7 @@ UNIT(LearnAND) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -79,7 +79,7 @@ UNIT(LearnNAND) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -119,7 +119,7 @@ UNIT(LearnOR) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -159,7 +159,7 @@ UNIT(LearnNOR) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -197,9 +197,9 @@ UNIT(LearnXOR) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 50'000, 0.25);
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
for (const auto & training_sample : training_sample_set_with_bias) {
std::vector<double> output;
@@ -233,7 +233,7 @@ UNIT(LearnNOT) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -271,7 +271,7 @@ UNIT(LearnX1) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);
@@ -309,7 +309,7 @@ UNIT(LearnX2) {
size_t num_examples = training_sample_set_with_bias.size();
size_t num_features = training_sample_set_with_bias[0].GetInputVectorSize();
size_t num_outputs = training_sample_set_with_bias[0].GetOutputVectorSize();
MLP my_mlp(num_features, num_outputs, 1, 2, false);
MLP my_mlp({ num_features, 2 ,num_outputs }, { "sigmoid", "linear" }, false);
//Train MLP
my_mlp.UpdateMiniBatch(training_sample_set_with_bias, 0.5, 500, 0.25);