AutoML with SmartCore

AutoML is Automated Machine Learning, referring to processes and methods to make machine learning more accessible for a general audience. This crate builds on top of the smartcore machine learning framework, and provides some utilities to quickly train and compare models.

Usage

Running the following: For instance, running the following:

fn main() {
  let data = smartcore::dataset::breast_cancer::load_dataset();
  let settings = automl::regression::Settings::default();
  let results = automl::regression::compare_models(data, settings);
  print!("{}", results);
}

Will output this:

┌───────────────────────────┬────────┬───────────┬──────────┐
│ Model                     │ R^2    │ MSE       │ MAE      │
╞═══════════════════════════╪════════╪═══════════╪══════════╡
│ Decision Tree Regression  │ 1.000  │ 1.638e-12 │ 5.531e-8 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ Random Forest Regression  │ 0.972  │ 6.626e-3  │ 2.830e-2 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ KNN Regression            │ 0.878  │ 2.851e-2  │ 5.624e-2 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ Linear Regression         │ 0.773  │ 5.309e-2  │ 1.813e-1 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ Ridge Regression          │ 0.772  │ 5.320e-2  │ 1.822e-1 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ Elastic Net               │ 0.385  │ 1.437e-1  │ 3.591e-1 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ LASSO                     │ 0.000  │ 2.338e-1  │ 4.675e-1 │
├───────────────────────────┼────────┼───────────┼──────────┤
│ Support Vector Regression │ -0.069 │ 2.500e-1  │ 5.000e-1 │
└───────────────────────────┴────────┴───────────┴──────────┘

Based on this output, you can then select the best model for the task.

Features

Currently this crate only includes AutoML features for regression and classification.

• Regression
  • Decision Tree Regression
  • KNN Regression
  • Random Forest Regression
  • Linear Regression
  • Rdige Regression
  • LASSO
  • Elastic Net
  • Support Vector Regression
• Classification
  • Random Forest Classification
  • Support Vector Classification
  • Logistic Regression
  • KNN Classification