LightweightMMM Documentation

LightweightMMM 🦇 is a lightweight Bayesian Marketing Mix Modeling (MMM) library that allows users to easily train MMMs and obtain channel attribution information. The library also includes capabilities for optimizing media allocation as well as plotting common graphs in the field.

It is built in python3 and makes use of Numpyro and JAX.

Installation

We have kept JAX as part of the dependencies to install when installing LightweightMMM, however if you wish to install a different version of JAX or jaxlib for specific CUDA/CuDNN versions see https://github.com/google/jax#pip-installation for instructions on installing JAX. Otherwise our installation assumes a CPU setup.

The recommended way of installing lightweight_mmm is through PyPi:

pip install --upgrade pip pip install lightweight_mmm

If you want to use the most recent and slightly less stable version you can install it from github:

pip install --upgrade git+https://github.com/google/lightweight_mmm.git

Model Documentation

• LightweightMMM Models.

Custom priors

• Custom priors in LightweigthMMM

API Documentation

• LightweightMMM
• LightweightMMM object
• Preprocessing / Scaling
• Optimize Media
• Plot
• Models
• Media Transforms
• Utils

FAQ

• Input
• Modelling
• Evaluation

Contribute

• Issue tracker: https://github.com/google/lightweight_mmm/issues

• Source code: https://github.com/google/lightweight_mmm/tree/main

Support

If you are having issues, please let us know by filing an issue on our issue tracker.

License

LightweightMMM is licensed under the Apache 2.0 License.

Indices and tables

• Index