BAyesian Model-Building Interface (Bambi) in Python

Bambi is a high-level Bayesian model-building interface written in Python. It works with two probabilistic programming frameworks, PyMC3 or PyStan, and is designed to make it extremely easy to fit Bayesian mixed-effects models common in biology, social sciences and other disciplines.

New Features

Bambi version 0.1.1 will be the final version supporting Python 2, but look forward to the forthcoming Bambi version 0.1.2!

Dependencies

Bambi is tested on Python 3.6 and depends on NumPy, Pandas, PyMC3, PyStan, Patsy and ArviZ (see requirements.txt for version information).

Installation

The latest release of Bambi can be installed using pip:

pip install bambi

Alternatively, if you want the bleeding edge version of the package, you can install from GitHub:

pip install git+https://github.com/bambinos/bambi.git

Usage

A simple fixed effects model is shown below as example.

from bambi import Model
import pandas as pd

# Read in a tab-delimited file containing our data
data = pd.read_table('my_data.txt', sep='\t')

# Initialize the model
model = Model(data)

# Fixed effects only model
results = model.fit('DV ~ IV1 + IV2', samples=1000, chains=4)

# Use ArviZ to plot the results
az.plot_trace(results)

# Key summary and diagnostic info on the model parameters
az.summary(results)

# Drop the first 100 samples (burn-in)
results_bi = results.sel(draw=slice(100, None))

For a more in-depth introduction to Bambi see our Quickstart or our set of example notebooks.

Contributing

We welcome contributions from interested individuals or groups! For information about contributing to Bambi, check out our instructions, policies, and guidelines here.

Contributors

See the GitHub contributor page.

Contents

• Getting Started
  • Quickstart
  • User Guide
    • Creating a model
      • Data format
    • Model specification
      • Formula-based specification
      • Incremental specification
      • Notes on fixed and random effects in Bambi
      • Coding of categorical variables
    • Fitting the model
      • Building the model
    • Alternative back-ends
    • Specifying priors
      • Different ways of specifying priors
      • A note on priors in Stan
      • Mapping priors onto terms
    • Generalized linear mixed models
      • Families
    • Results
    • Plotting
    • Summarizing
    • Accessing back-end objects
• Examples
  • Bayesian Logistic Regression Example
    • Load and examine American National Election Studies (ANES) data
    • Build clinton_model
    • Run Inference
    • Spaghetti plot of model predictions
  • Bayesian/Frequentist Tutorial
    • Generate t-test data
      • Frequentist
      • Bayesian
    • Generate multi-level regression data
      • Frequentist multi-level model
      • Bayesian multi-level model
  • Bayesian Multiple Regression Example
    • Load and examine Eugene-Springfield community sample data
    • Specify model and examine priors
    • Examine the model results
    • Summarize effects on partial correlation scale
    • Relative importance: Which predictors have the strongest effects (defined in terms of partial \(\eta^2\))?
  • Bayesian Workflow Example (Police Officer’s Dilemma)
    • Load and examine data
    • Fit response rate models
      • Random subject effects only
      • Random stimulus effects
    • Fit response models
  • Bayesian Workflow Example (Strack RRR Analysis Replication)
    • Reading in the data
    • Reshaping the data
    • Fitting the model
    • Inspecting results
    • Expanding the model
    • And the answer is…
• API Reference
  • bambi.models
  • bambi.priors

Indices

• Index

• Module Index

Getting Started