In greta-dev/greta: Simple and Scalable Statistical Modelling in R

Improper flat prior

A simple, one-variable Bayesian linear regression model that uses flat priors for the coefficients. A flat prior using variable puts an unbounded uniform distribution on the parameter. With unconstrained flat priors, the posterior will be proportional to the likelihood and the MAP will correspond to the MLE. Flat priors are usually chosen when there is little knowledge about the parameters available.

# variables & priors
int  <- variable()
coef <- variable()
sd   <- cauchy(0, 3, truncation = c(0, Inf))

# linear predictor
mu <- int + coef * attitude$complaints

# observation model
distribution(attitude$rating) <- normal(mu, sd)