Bayesian inference sandbox in C
u know wat it is
make bayes && ./bayes
Basically, we're adding data points to an array of fixed length. Each time we add a data point we update the
kappa of our gamma-normal distribution. At the end (after loading all the data) we print out the final
mu our model has converged on.
You can experiment adding many data points that are close together, then a few outliers, to see how/to what extent the gamma-normal is resilient to rogue data.