vistan 0.0.0.2

Library to run VI algorithms on Stan models.

vistan

vistan is a simple library to run variational inference algorithms on Stan models.

Features

• Initialization: Laplace's method to initialize full-rank Gaussian
• Gradient Estimators: Total-gradient, STL, DReG, closed-form entropy
• Variational Families: Full-rank Gaussian, Diagonal Gaussian, RealNVP
• Objectives: ELBO, IW-ELBO
• IW-sampling: Posterior samples using importance weighting

Installation

pip install vistan

Usage

Meanfield VI

code = """
data {
  int<lower=0> J;         // number of schools
  real y[J];              // estimated treatment effects
  real<lower=0> sigma[J]; // standard error of effect estimates
}
parameters {
  real mu;                // population treatment effect
  real<lower=0> tau;      // standard deviation in treatment effects
  vector[J] eta;          // unscaled deviation from mu by school
}
transformed parameters {
  vector[J] theta = mu + tau * eta;        // school treatment effects
}
model {
  target += normal_lpdf(eta | 0, 1);       // prior log-density
  target += normal_lpdf(y | theta, sigma); // log-likelihood
}
"""

data = {"J": 8,
                "y": [28,  8, -3,  7, -1,  1, 18, 12],
                "sigma": [15, 10, 16, 11,  9, 11, 10, 18]}

posterior, model, results = vistan.infer(code = code, data = data) # runs Meanfield VI by default

samples = posterior.sample(1000)
for i in range(samples['eta'].shape[1]):
    plt.plot(samples["eta"][:,i], label = "eta[i]")
plt.show()

Gaussian VI

hyperparams = vistan.hyper_params(method = 'gaussian')

posterior, model, results = vistan.infer(code = code, data = data, 
                        hyperparams = hyperparams, verbose = True)

samples = posterior.sample(1000)
for i in range(samples['eta'].shape[1]):
    plt.plot(samples["eta"][:,i], label = "eta[i]")
plt.show()

Flow-based VI

hyperparams = vistan.hyper_params(method = 'flows')

posterior, model, results = vistan.infer(code = code, data = data, 
                        hyperparams = hyperparams, verbose = True)

samples = posterior.sample(1000)
for i in range(samples['eta'].shape[1]):
    plt.plot(samples["eta"][:,i], label = "eta[i]")
plt.show()

ADVI

hyperparams = vistan.hyperparams(method = 'advi')

posterior, model, results = vistan.infer(code = code, data = data, 
                                hyperparams = hyperparams, verbose = True)

samples = posterior.sample(1000)
for i in range(samples['eta'].shape[1]):
    plt.plot(samples["eta"][:,i], label = "eta[i]")
plt.show()

Custom

hyperparams = vistan.hyperparams(   method = 'custom', 
                                    vi_family = "gaussian",
                                    M_training = 10,
                                    grad_estimator = "DReG",
                                    LI = True)

posterior, model, results = vistan.infer(code = code, data = data, 
                                hyperparams = hyperparams, verbose = True)

samples = posterior.sample(1000)
for i in range(samples['eta'].shape[1]):
    plt.plot(samples["eta"][:,i], label = "eta[i]")
plt.show()

Limitations

• We will currently only support inference on all latent parameters in the model