greta-dev/greta

simple and scalable statistical modelling in R

What it solves

It provides a way to write and fit statistical models directly in R, offering an alternative to tools like BUGS, JAGS, and Stan. It is designed to scale to massive datasets and be easily extendable.

How it works

Greta uses Google's TensorFlow and TensorFlow Probability under the hood to handle the underlying computations and model fitting via MCMC (Markov Chain Monte Carlo) and optimization.

Who it’s for

Statisticians and data scientists using R who need to build custom statistical models that can scale to large datasets.

Highlights

  • Written directly in R
  • Scales to massive datasets
  • Uses TensorFlow and TensorFlow Probability for the backend
  • Supports MCMC and optimization for model fitting

Related

  • Project
    topepo/caret

    A comprehensive R package for training and plotting classification and regression models, providing a unified interface for machine learning tasks.

  • Project
    yrosseel/lavaan

    An open-source R package for latent variable analysis, supporting structural equation modeling, path analysis, and factor analysis.

  • Project
    tensorflow/gnn

    A TensorFlow library for building and scaling Graph Neural Networks, featuring tools for heterogeneous graph representation and distributed graph sampling.

  • Project
    tensorflow/probability

    A library for probabilistic reasoning and statistical analysis that integrates probabilistic methods with deep networks using TensorFlow or JAX.

  • Project
    datalorax/equatiomatic

    An R package that automatically converts fitted statistical models into LaTeX equations, reducing the manual effort of writing mathematical formulas for research papers.