bsmart.scans.AffineMC

Affine MCMC scan using the Goodman & Weare algorithm.

This is an excellent robust MCMC algorithm for larger numbers of dimensions.

The chain, and automatic plots, are stored in a Results subdirectory. Results are updated after each pass.

The relevant scan-specific settings are:

"Setup" : {
    "Steps": "int, Number of steps",
    "Walkers" : "int, Number of walkers, default 10 * number of variables"
}

To store observable information inside the results file, it is necessary to set:

"Setup" : {
    "store_points_in_memory": "True",
    "store_invalid_points": "True"
}

Information

BSMArt Name: AffineMC

Requires:

  • matplotlib

  • numpy

  • corner

  • pandas

  • seaborn

Settings:

  • Steps: int, Number of steps

  • Walkers: int, Number of walkers, default 10 * number of variables

class bsmart.scans.AffineMC.NewScan(inputs, log)[source]

Bases: Scan

Scanner class for Affine MCMC Scans

initialise()[source]

method to allow the user scan to overload run settings etc during the initialisation process

postprocess(Point, observables, data_point, temp_dir, log, lock=None)[source]

return the likelihood; we won’t get this far if the point failed to be generated

run()[source]

Affine MCMC: Uses Goodman & Weare Affine Invariant Ensemble Sampler

run_batch(thetas)[source]