David Augustin

Create a functional testing module with tests for all methods

5

The module will contain the functional tests in pints, e.g. one file for each sampler and each optimiser. We start with the HarrioBardenerACMC.

Change-point tests

5

Closes #1294 .

I1137 horowitz langevin mcmc

17

Closes #1137 The algorithm is largely the same as the existing a `pints.HamiltonianMCMC`. I made references in the ask and tell methods where the code has been adapted. The other...

I was just discussing with @ben18785 that it may be beneficial to make the sampling and optimisation methods more accessible by providing some intuition on the algorithms in the docstrings...

Compute pointwise log-likelihood for each observation

7

ArviZ provides a simple API to compute the LOO or WAIC for performance assessment of models, see https://arviz-devs.github.io/arviz/api/generated/arviz.waic.html. What this would require however is the pointwise log-likelihood scores of the...

Optimise performance of NUTS

9

The current implementation of the `pints.NoUTurnMCMC` is at least 10 times slower than `pints.HamiltonianMCMC` on a problem I tried (5 dimensional parameter space, default settings), which probably renders this implementation...

Store energies for HMC samplers

1

It seems quite established to compare the distribution of energy transition to the overall energy distribution in a chain to determine whether the momenta / mass matrix where chosen appropriately...

Seeding of priors

4

At the moment it's very easy to sample from priors, but we do not have any way to seed these random samples. I find seeding especially useful when exploring the...

Move MCMCSummary to MCMC diagnostics in the docs

5

After a while I found the `pints.MCMCSummary` object with the MCMC samplers in the documentation, and I was somehow expecting it to be with the diagnostics. Is this just me...

Modify Rhat to compute rank-normalised rhat

3

This issue partially addresses #1063 . I'd like to first address rank-normalising rhat, because this just involves - [ ] Computing the rank statistic of the pooled samples - [...