Using PolyChron

Installing PolyChron

For installation instructions, see Getting Started.

Launching PolyChron

Once installed, polychron can be launched from the command line.

# via the executable script
polychron
# or by running the installed polychron module
python3 -m polychron

On launch, you should see a window as below.

The version of polychron may be checked using --version

polychron --version

For full command line options, see -h / --help

polychron --help

The PolyChron Workflow

Usage of polychron can be split into a 4 stage workflow:

graph LR
    g0[Project loading];
    g1[Prior elicitation];
    g2[Posterior inference];
    g3[Post-MCMC analysis];
    g0 --> g1;
    g1 --> g2;
    g2 --> g3;

Project Loading

Project Loading allows the user to create or select a project and model to manipulate and interrogate.

See Project Loading for more information.

Prior Elicitation

Prior Elicitation involves building and manipulating a model with stratigraphic relationships, context grouping, group orderings and dating information to form the stratigraphic graph.

The stratigraphic graph can then be combined with additional information, such as identify residual & intrusive contexts; and the types of group relationships, to form the chronological graph.

See Prior Elicitation for more information.

Posterior Inference

Once the user has loaded the chronological graph for a given model, posterior estimates can be computed using a Monte Carlo Markov Chain (MCMC) algorithm.

See Posterior Inference for more information.

Post-MCMC Analysis

Finally, once the posterior inference is complete and the minimum required number of MCMC samples has been met, users can interrogate the MCMC results including marginal HPD intervals and posterior probability density plots.

See Post-MCMC Analysis for more information.

Configuration

PolyChron has several application-wide configuration parameters. See Configuration for more details.