Intro

DartML is an application that empowers users to analyze CSV data and build machine learning models without writing any code. It’s designed to make machine learning accessible to everyone, regardless of their programming experience.

DartML home page

Features

1. Data Upload and Preparation

  • Specify sample size
  • Choose between train/test split and cross-validation
  • Define shuffle and stratify options

Data Upload

2. Data Exploration

Easily explore both numerical and categorical data with built-in visualizations.

Numerical data exploration Categorical data exploration

3. Model Building

  • Choose the target column
  • Select the problem type (can be automatically detected)
  • Define metric (e.g., accuracy, F1 score, etc.)
  • Choose algorithms
  • Specify total training time

Model Building

4. Model Evaluation

All tables and plots are updated in real-time during training!

  • Leaderboard
  • Performance Boxplot
  • Feature Importance (eXplainable AI)
  • Spearman Correlation of Models
  • Training Logs

Leaderboard Performance Boxplot Features Importance Spearman Correlation Logs

5. Detailed Reports

Download comprehensive reports with all trained models and detailed information, including SHAP and dtreeviz visualizations.

SHAP values dtreeviz visualization

Resources