/mutagen

Source code and slides for an end-to-end tidymodels demo

Primary LanguageHTML

A diagram representing the cycle of model development. Clockwise, the diagram transitions from collecting and cleaning data to training models to deploying models. Beside each step is a hex sticker for tidyverse, tidymodels, and vetiver, respectively.

An End-to-End Data Science Workflow

This repository contains source code demonstrating a tidymodels machine learning pipeline for prediction of drug mutagenicity. See the following links to learn more about different components of the analysis:

Blog Post

A post on the Posit Blog titled "Pharmaceutical machine learning with tidymodels and Posit Connect," where we walk through the model development process at a high level.

Model Training Notebook

Annotated source code to propose, fit, and evaluate a number of machine learning models.

Slides

Materials for a presentation at the Posit Team Demo series.

Repository Structure

  • source/ contains scripts to propose, fit, and evaluate a number of machine learning models.
  • slides/ contains Quarto source code for the linked slides.
  • data/ contains summarized output from these scripts, for use in the write-up of the process.
  • data-raw/ contains raw output from these scripts and is not tracked, as these files are quite large.

index.html contains unbranded source for the post on the Posit Blog.