Repository files navigation

fastml is a streamlined R package designed to simplify the training, evaluation, and comparison of multiple machine learning models. It offers comprehensive data preprocessing, supports a wide range of algorithms with hyperparameter tuning, and provides performance metrics alongside visualization tools to facilitate efficient and effective machine learning workflows.

• Comprehensive Data Preprocessing: Handle missing values, encode categorical variables, and apply various scaling methods with minimal code.
• Support for Multiple Algorithms: Train a wide array of machine learning models including XGBoost, Random Forest, SVMs, KNN, Neural Networks, and more.
• Hyperparameter Tuning: Customize and automate hyperparameter tuning for each algorithm to optimize model performance.
• Performance Evaluation: Evaluate models using metrics like Accuracy, Kappa, Sensitivity, Specificity, Precision, F1 Score, and ROC AUC.
• Visualization Tools: Generate comparison plots to visualize and compare the performance of different models effortlessly.
• Easy Integration: Designed to integrate seamlessly into your existing R workflows with intuitive function interfaces.

You can install the latest stable version offastml from CRAN using:

install.packages("fastml")

You can install all dependencies (additional models) using:

# install all dependencies - recommendedinstall.packages("fastml",dependencies=TRUE)

For the development version, install directly from GitHub using the devtools package:

# Install devtools if you haven't alreadyinstall.packages("devtools")# Install fastml from GitHubdevtools::install_github("selcukorkmaz/fastml")

Here's a simple workflow to get you started with fastml:

library(fastml)# Example datasetdata(iris)iris<-iris[iris$Species!="setosa", ]# Binary classificationiris$Species<-factor(iris$Species)# Train modelsmodel<- fastml(data=iris,label="Species")# View model summarysummary(model)

fastml supports both grid search and Bayesian optimization through thetuning_strategy argument. Use"grid" for a regular parameter grid or"bayes" for Bayesian hyperparameter search. Thetuning_iterationsparameter controls the number of iterationsonly whentuning_strategy = "bayes" and is ignored otherwise.