tidyML

Machine learning pipelines are too often built from spaghetti. This package provides utilities to consolidate logic and minimize duplication, which helps you write maintainable code.

Modules

• DataMediator

• ExperimentTracker

  • NeptuneExperimentTracker
  • WandbExperimentTracker

• HyperparameterOptimizer

  • BayesianOptimizer
  • GaussianProcessRegressor
  • RegressorCollection

Environment

• Python 3.9 or greater

Development quickstart

0. Install Python, open a terminal in the tidyML directory

1. Install dependencies with pip install -r requirements.txt