VectorForgeML Documentation

Usage guides, API references, and examples for the VectorForgeML machine learning framework. Built on C++ for maximum performance in R.

Quick Start Guide

Get up and running with a simple classification pipeline in VectorForgeML.

First, install VectorForgeML, then build your first pipeline:

library(VectorForgeML)

# Load data
data(iris)

# Create pipeline
pipe <- Pipeline(steps = list(
    c("scaler", StandardScaler()),
    c("model", SoftmaxRegression())
))

# Train
pipe$fit(iris[, 1:4], iris[, 5])

API Reference

Explore the available modules and algorithms in VectorForgeML.

Models

Preprocessing

StandardScaler

Feature Scaling — Zero Mean, Unit Variance

MinMaxScaler

Feature Scaling — Range Normalization

Pipeline

End-to-End ML Workflow

LabelEncoder

Categorical Encoding

OneHotEncoder

Binary Feature Expansion

ColumnTransformer

Column-wise Transformations

train_test_split

Data Splitting Utility

drop_constant_columns

Feature Selection

Evaluation Metrics

accuracy_score

Classification Accuracy

f1_score

Harmonic Mean of Precision & Recall

confusion_matrix

Classification Error Analysis

r2_score

Coefficient of Determination

mse / rmse

Regression Error Metrics

precision_score / recall_score

Classification Precision & Recall