Engage with the ML in Java Community!

[adityakrmishra]

👋 Welcome to ML in Java!

We’re using Discussions as a place to connect with other members of our community. This is the perfect spot to share ideas, ask questions, and collaborate on our "ML in Java" project. Our goal is to create a collection of machine learning algorithms and utilities implemented in Java, making it easier to build and evaluate machine learning models efficiently.

Project Overview

The project is organized into the following components:

• DataLoader: Utility class for loading and preprocessing data.
• MathUtils: Contains mathematical utility functions.
• LinearRegression: Implementation of linear regression.
• DecisionTree: Implementation of a decision tree classifier.
• GradientDescent: Gradient descent optimization algorithm.
• KMeansClustering: Implementation of the k-means clustering algorithm.
• CrossValidation: Cross-validation utilities for model evaluation.
• Metrics: Evaluation metrics for regression, classification, and clustering.

Features

• DataLoader:

  • loadCSV(String filePath): Loads data from a CSV file.

• MathUtils:

  • mean(double[] values): Calculates the mean of an array of doubles.
  • standardDeviation(double[] values): Calculates the standard deviation of an array of doubles.

• LinearRegression:

  • fit(double[][] X, double[] y): Fits the linear regression model.
  • predict(double[][] X): Predicts target values using the fitted model.

• DecisionTree:

  • fit(double[][] X, double[] y): Fits the decision tree classifier.
  • predict(double[][] X): Predicts target values using the fitted classifier.
  • prune(int maxDepth): Prunes the decision tree to the specified maximum depth.

• GradientDescent:

  • fit(double[][] X, double[] y): Fits the model using gradient descent.
  • predict(double[][] X): Predicts target values using the fitted model.
  • fitPolynomial(double[][] X, double[] y, int degree): Performs polynomial regression.

• KMeansClustering:

  • fit(double[][] X): Fits the k-means clustering model.
  • predict(double[][] X): Predicts cluster labels for the input features.
  • getCentroids(): Returns the centroids of the clusters.
  • calculateSSE(): Calculates the sum of squared errors for the clusters.

• CrossValidation:

  • kFoldCrossValidation(Model model, double[][] X, double[] y, int k): Performs k-fold cross-validation.
  • stratifiedKFoldCrossValidation(Model model, double[][] X, double[] y, int k): Performs stratified k-fold cross-validation.
  • leaveOneOutCrossValidation(Model model, double[][] X, double[] y): Performs leave-one-out cross-validation.
  • randomSplitCrossValidation(Model model, double[][] X, double[] y, double testSize): Performs random split cross-validation.
  • monteCarloCrossValidation(Model model, double[][] X, double[] y, int numSplits, double testSize): Performs Monte Carlo cross-validation.
  • bootstrapCrossValidation(Model model, double[][] X, double[] y, int numSamples): Performs bootstrap cross-validation.

• Metrics:

  • meanSquaredError(double[] yTrue, double[] yPred): Calculates mean squared error for regression problems.
  • rootMeanSquaredError(double[] yTrue, double[] yPred): Calculates root mean squared error for regression problems.
  • meanAbsoluteError(double[] yTrue, double[] yPred): Calculates mean absolute error for regression problems.
  • rSquared(double[] yTrue, double[] yPred): Calculates R-squared for regression problems.
  • accuracy(double[] yTrue, double[] yPred): Calculates accuracy for classification problems.
  • precision(double[] yTrue, double[] yPred): Calculates precision for classification problems.
  • recall(double[] yTrue, double[] yPred): Calculates recall for classification problems.
  • f1Score(double[] yTrue, double[] yPred): Calculates F1 score for classification problems.
  • confusionMatrix(double[] yTrue, double[] yPred): Calculates the confusion matrix for classification problems.
  • aucRoc(double[] yTrue, double[] yPred): Calculates the AUC-ROC score for binary classification problems.
  • matthewsCorrelationCoefficient(double[] yTrue, double[] yPred): Calculates the Matthews correlation coefficient for binary classification problems.
  • precisionRecallF1MultiClass(double[] yTrue, double[] yPred): Calculates precision, recall, and F1 score for each class in a multi-class classification problem.
  • adjustedRandIndex(double[] labelsTrue, double[] labelsPred): Computes the Adjusted Rand Index for clustering evaluation.
  • silhouetteScore(double[][] X, double[] labels): Computes the Silhouette Score for clustering evaluation.
  • homogeneityScore(double[] labelsTrue, double[] labelsPred): Computes the Homogeneity score for clustering evaluation.
  • completenessScore(double[] labelsTrue, double[] labelsPred): Computes the Completeness score for clustering evaluation.
  • vMeasureScore(double[] labelsTrue, double[] labelsPred): Computes the V-Measure score for clustering evaluation.
  • mutualInformationScore(double[] labelsTrue, double[] labelsPred): Computes the Mutual Information score for clustering evaluation.
  • normalizedMutualInformationScore(double[] labelsTrue, double[] labelsPred): Computes the Normalized Mutual Information score for clustering evaluation.
  • randIndex(double[] labelsTrue, double[] labelsPred): Computes the Rand Index for clustering evaluation.
  • fowlkesMallowsIndex(double[] labelsTrue, double[] labelsPred): Computes the Fowlkes-Mallows Index for clustering evaluation.
  • jaccardIndex(double[] labelsTrue, double[] labelsPred): Computes the Jaccard Index for clustering evaluation.
  • meanSquaredLogarithmicError(double[] yTrue, double[] yPred): Computes the Mean Squared Logarithmic Error for regression problems.
  • explainedVariance(double[] yTrue, double[] yPred): Computes the Explained Variance Score for regression problems.
  • medianAbsoluteError(double[] yTrue, double[] yPred): Computes the Median Absolute Error for regression problems.
  • hammingLoss(double[] yTrue, double[] yPred): Computes the Hamming Loss for classification problems.

Usage

To use these classes and methods, you can import the respective classes into your Java code and create instances of the models or call the utility methods as needed.

Join the Discussion

We hope that you:

• Ask questions you're wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and be open-minded. Remember that this is a community we build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

Looking forward to connecting with all of you!

Best regards,
[Your Name]