Naaturel/ANN-framework
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Minor changes

Browse Source
This commit is contained in:
Laurent
2026-03-27 12:40:00 +01:00
parent 572e5c7484
commit 7fb4a7c057
3 changed files with 29 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/main/java/com/naaturel/ANN/Main.java
View File

@@ -38,7 +38,7 @@ public class Main {
Layer layer = new Layer(neurons); Layer layer = new Layer(neurons);
Network network = new Network(List.of(layer)); Network network = new Network(List.of(layer));
Trainer trainer = new AdalineTraining(); Trainer trainer = new GradientDescentTraining();
trainer.train(network, dataset); trainer.train(network, dataset);
} }

42
src/main/java/com/naaturel/ANN/domain/model/training/TrainingPipeline.java
View File

@@ -81,25 +81,7 @@ public class TrainingPipeline {
System.out.printf("[Training finished in %.3fs]", (end-start)/1000.0); System.out.printf("[Training finished in %.3fs]", (end-start)/1000.0);
} }
AtomicInteger neuronIndex = new AtomicInteger(0); if(this.visualization) this.visualize(ctx);
ctx.model.forEachNeuron(n -> {
List<Float> weights = new ArrayList<>();
n.forEachSynapse(syn -> weights.add(syn.getWeight()));
float b = weights.get(0);
float w1 = weights.get(1);
float w2 = weights.get(2);
this.visualizer.addEquation("boundary_" + neuronIndex.getAndIncrement(), w1, w2, b, -3, 3);
});
int i = 0;
for(DataSetEntry entry : ctx.dataset){
List<Input> inputs = entry.getData();
this.visualizer.addPoint("p"+i, inputs.get(0).getValue(), inputs.get(1).getValue());
this.visualizer.addPoint("p"+i, inputs.get(0).getValue()+0.01F, inputs.get(1).getValue()+0.01F);
i++;
}
this.visualizer.build();
} }
private void executeSteps(TrainingContext ctx){ private void executeSteps(TrainingContext ctx){
@@ -123,4 +105,26 @@ public class TrainingPipeline {
ctx.epoch += 1; ctx.epoch += 1;
} }
private void visualize(TrainingContext ctx){
AtomicInteger neuronIndex = new AtomicInteger(0);
ctx.model.forEachNeuron(n -> {
List<Float> weights = new ArrayList<>();
n.forEachSynapse(syn -> weights.add(syn.getWeight()));
float b = weights.get(0);
float w1 = weights.get(1);
float w2 = weights.get(2);
this.visualizer.addEquation("boundary_" + neuronIndex.getAndIncrement(), w1, w2, b, -3, 3);
});
int i = 0;
for(DataSetEntry entry : ctx.dataset){
List<Input> inputs = entry.getData();
this.visualizer.addPoint("p"+i, inputs.get(0).getValue(), inputs.get(1).getValue());
this.visualizer.addPoint("p"+i, inputs.get(0).getValue()+0.01F, inputs.get(1).getValue()+0.01F);
i++;
}
this.visualizer.build();
}
} }

8
src/main/java/com/naaturel/ANN/implementation/training/GradientDescentTraining.java
View File

@@ -12,6 +12,7 @@ import com.naaturel.ANN.implementation.gradientDescent.SquareLossStrategy;
import com.naaturel.ANN.implementation.simplePerceptron.SimpleDeltaStrategy; import com.naaturel.ANN.implementation.simplePerceptron.SimpleDeltaStrategy;
import com.naaturel.ANN.implementation.simplePerceptron.SimplePredictionStrategy; import com.naaturel.ANN.implementation.simplePerceptron.SimplePredictionStrategy;
import com.naaturel.ANN.implementation.training.steps.*; import com.naaturel.ANN.implementation.training.steps.*;
import com.naaturel.ANN.infrastructure.graph.GraphVisualizer;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.List; import java.util.List;
@@ -27,7 +28,7 @@ public class GradientDescentTraining implements Trainer {
GradientDescentTrainingContext context = new GradientDescentTrainingContext(); GradientDescentTrainingContext context = new GradientDescentTrainingContext();
context.dataset = dataset; context.dataset = dataset;
context.model = model; context.model = model;
context.learningRate = 0.0011F; context.learningRate = 0.0005F;
context.correctorTerms = new ArrayList<>(); context.correctorTerms = new ArrayList<>();
List<TrainingStep> steps = List.of( List<TrainingStep> steps = List.of(
@@ -38,7 +39,7 @@ public class GradientDescentTraining implements Trainer {
); );
new TrainingPipeline(steps) new TrainingPipeline(steps)
.stopCondition(ctx -> ctx.globalLoss <= 0.08F || ctx.epoch > 500) .stopCondition(ctx -> ctx.globalLoss <= 0.00F || ctx.epoch > 50000)
.beforeEpoch(ctx -> { .beforeEpoch(ctx -> {
GradientDescentTrainingContext gdCtx = (GradientDescentTrainingContext) ctx; GradientDescentTrainingContext gdCtx = (GradientDescentTrainingContext) ctx;
gdCtx.globalLoss = 0.0F; gdCtx.globalLoss = 0.0F;
@@ -49,8 +50,9 @@ public class GradientDescentTraining implements Trainer {
context.globalLoss /= context.dataset.size(); context.globalLoss /= context.dataset.size();
new GradientDescentCorrectionStrategy(context).apply(); new GradientDescentCorrectionStrategy(context).apply();
}) })
.withVerbose(true) //.withVerbose(true)
.withTimeMeasurement(true) .withTimeMeasurement(true)
.withVisualization(false, new GraphVisualizer())
.run(context); .run(context);
} }