tmmc-lnpy

Overview

A package to analyze $\ln \Pi(N)$ data from Transition Matrix Monte Carlo simulation. The main output from TMMC simulations, $\ln \Pi(N)$, provides a means to calculate a host of thermodynamic properties. Moreover, if $\ln \Pi(N)$ is calculated at a specific chemical potential, it can be reweighted to provide thermodynamic information at a different chemical potential

Features

tmmc-lnpy provides a wide array of routines to analyze $\ln \Pi(N)$. These include:

• Reweighting to arbitrary chemical potential
• Segmenting $\ln \Pi(N)$ (to identify unique phases)
• Containers for interacting with several values of $\ln \Pi(N)$ in a vectorized way.
• Calculating thermodynamic properties from these containers
• Calculating limits of stability, and phase equilibrium

Status

This package is actively used by the author. Please feel free to create a pull request for wanted features and suggestions!

Example usage

Note that the distribution name tmmc-lnpy is different than the import name lnpy due to name clashing on pypi.

>>> import numpy as np
>>> import lnpy
>>> import lnpy.examples

>>> ref = lnpy.examples.load_example_lnpimasked("lj_sub")

>>> phase_creator = lnpy.PhaseCreator(nmax=1, ref=ref)
>>> build_phases = phase_creator.build_phases_mu([None])
>>> collection = lnpy.lnPiCollection.from_builder(
...     lnzs=np.linspace(-10, 3, 5), build_phases=build_phases
... )


# Collections are like pandas.Series
>>> collection
<class lnPiCollection>
lnz_0   phase
-10.00  0        [-10.0]
-6.75   0        [-6.75]
-3.50   0         [-3.5]
-0.25   0        [-0.25]
 3.00   0          [3.0]
dtype: object


# Access xarray backend for Grand Canonical properties with `xge` accessor
>>> collection.xge.betaOmega()
<xarray.DataArray 'betaOmega' (lnz_0: 5, phase: 1)> Size: 40B
array([[-2.3245e-02],
       [-6.0370e-01],
       [-1.8552e+02],
       [-1.5447e+03],
       [-2.9580e+03]])
Coordinates:
  * lnz_0    (lnz_0) float64 40B -10.0 -6.75 -3.5 -0.25 3.0
  * phase    (phase) int64 8B 0
    beta     float64 8B 1.372
    volume   float64 8B 512.0
Attributes:
    dims_n:         ['n_0']
    dims_lnz:       ['lnz_0']
    dims_comp:      ['component']
    dims_state:     ['lnz_0', 'beta', 'volume']
    dims_rec:       ['sample']
    standard_name:  grand_potential
    long_name:      $\beta \Omega(\mu,V,T)$

Installation

Use one of the following

pip install tmmc-lnpy

or

conda install -c conda-forge tmmc-lnpy

Documentation

See the documentation for a look at tmmc-lnpy in action.

What's new?

See changelog.

License

This is free software. See LICENSE.

Related work

This package is used for with thermoextrap to analyze thermodynamically extrapolated macro state probability distributions.

Contact

The author can be reached at wpk@nist.gov.

Credits

This package was created using Cookiecutter with the usnistgov/cookiecutter-nist-python template.