Closed drbenvincent closed 5 years ago
So far we visualise the posterior by plotting the univariate marginal histograms.
But it would be better to do a corner plot, so we can see not only these univariate marginals, but also the bivariate joint distributions.
This would be useful in diagnosing any issues with inference or the models.
That's basically that. Less than 50 lines. More improvements to come, but we are done with this issue
So far we visualise the posterior by plotting the univariate marginal histograms.
But it would be better to do a corner plot, so we can see not only these univariate marginals, but also the bivariate joint distributions.
This would be useful in diagnosing any issues with inference or the models.