GitHub - sarameller/edux: EDUX is a developer friendly Java library for machine learning educational tasks

EDUX is a user-friendly library for solving problems with a machine learning approach.

Goal

The main goal of this project is to create a user-friendly library for solving problems using a machine learning approach. The library is designed to be easy to use, enabling the solution of problems with just a few lines of code.

Installation

To install, include the library as a dependency in your Java project file.

Gradle

 implementation 'io.github.samyssmile:edux:1.0.4'

Maven

  <dependency>
     <groupId>io.github.samyssmile</groupId>
     <artifactId>edux</artifactId>
     <version>1.0.4</version>
 </dependency>

Usage - Neural Network

You need to implement a IDataProvider Interface to provide the data for the network. Example for the IRIS-Dataset.

  public class IrisProvider implements IDataProvider<Iris>

Create a Neural Network Configuration.

    var neuralNetworkConfiguration = new Configuration(
                                         inputSize, 
                                         hiddenLayersSize, 
                                         outputSize, 
                                         learningRate,
                                         epochs,
                                         ActivationFunction.LEAKY_RELU, // Activation function of the hidden layers
                                         ActivationFunction.SOFTMAX, // Activation function of the output layer
                                         LossFunction.CATEGORICAL_CROSS_ENTROPY);// Loss function of the network

Create a Neural Network with the configuration and data from provider. Finaly train the network.

   var neuralNetwork = new NeuralNetwork(features, labels, testFeatures, testLabels, neuralNetworkConfiguration);
   neuralNetwork.train();

You can predict the output with the predict method.

   var prediction = neuralNetwork.predict(yourInputYouWantToPredict);

K Nearest Neighbors

Its also possible to use the K Nearest Neighbors algorithm.

KnnClassifier knnClassifier = new KnnClassifier(1, labeledPoints);

a full working example can be found in

package knn.example.de.nexent.edux.KnnExample.java

Implement IDataProvider Interface to get your data.
Transform your data to LabeledPoints.
Create a KnnClassifier with the number of neighbors and the labeled points.

You can evaluate the accuracy of the classifier with the evaluate method.

knnClassifier.evaluate(testLabeledPoints);

output
[main] INFO  knn.ml.de.nexent.edux.KnnClassifier - Accuracy: 98.33333333333333%

To classify a single record use the classify method.

knnClassifier.classify(...);

Decision Tree

A full working example can be found in

package decisiontree.example.de.nexent.edux.DecisionTreeExample.java

Example Output for IRIS Datenset

de.edux.ml.decisiontree.DecisionTree - Accuracy: 95,0000%
Feature Importance: [0.4444444444444444, 0.0, 0.4291223175466741, 0.42559077319203725]

Features

The library currently supports:

Multilayer Perceptron
K Nearest Neighbors
Descision Tree
Support Vector Machine (progress)
Random Forest (progress)

Contributions

Contributions are warmly welcomed! If you find a bug, please create an issue with a detailed description of the problem. If you wish to suggest an improvement or fix a bug, please make a pull request.