Skip to main content

Bayesian optimization for iterated multi-armed bandit experiments.

Project description

# Dillinger: Deadly accurate multi-armed bandits

Dillinger is a guide to using Bayesian optimization to select new actions for multi-armed bandits. The core of the project is a **Gaussian Process** class that can be fit to observations from multi-armed bandits. To facilitate demonstration, the package also has the following features: a data generator that simulates LTV of customers based on a price sensitivity curve, and an implementation of the Softmax bandit algorithm.

This project is still very much under construction, as I'm adapting an existing project to make it more useable and accessible to those interested in applying Bayesian optimization to A/B tests or multi-armed bandit experiments.

See `demos\` for examples of how to use this package.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dillinger-1.0.0.dev1.tar.gz (8.9 kB view hashes)

Uploaded source

Built Distribution

dillinger-1.0.0.dev1-py3-none-any.whl (12.0 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page