This repository includes a number of python scripts that are useful for working with quantum Monte Carlo data generated via the Del Maestro group PIMC code which is located at https://code.delmaestro.org.
For now, we have not uploaded to pypi but the scripts can be installed directly from git:
pip install --upgrade git+https://github.com/DelMaestroGroup/pimcscripts.git#egg=pimcscripts
This will install the base library pimcscripts which includes the modules
pimcscripts.pimchelp and pimcscripts.MCStat as well as a number of very
useful helper scripts located in ./bin. These should be installed to your path and include documentation that can found via:
script_name.py --help
If you are upgrading after a change, it might be useful to use:
pip install --upgrade --no-deps --force-reinstall git+https://github.com/DelMaestroGroup/pimcscripts.git#egg=pimcscripts
Below we describe a few useful ones.
usage: pimcave.py [-h] [-s SKIP] [-l HEADER_LINES] [-r] file [file ...]
Generates averages from pimc output data.
positional arguments:
file File or files to average.
optional arguments:
-h, --help show this help message and exit
-s SKIP, --skip SKIP How many input lines should we skip? [default: 0]
-l HEADER_LINES, --header_lines HEADER_LINES
How many header lines to skip?
-r, --repeated_header
deal with duplicate headers
pimcplot
Description:
Performs a cumulative average plot of raw Monte Carlo data
Usage:
pimcplot.py [options] [--legend=<label>...] --estimator=<name> <file>...
pimcplot.py -h | --help
Options:
-h, --help Show this screen.
--estimator=<name>, -e <name> The estimator to be plotted.
--skip=<n>, -s <n> Number of measurements to be skipped [default: 0].
--period=<m>, -p <m> The period of the average window [default: 50].
--truncateid=<t>, -t <t> Truncate PIMCID to last <t> characters [default: 0].
--legend=<label>, -l <label> A legend label
--period=<m>, -p <m> The period of the average window [default: 50].
--error=<units>, -d Size of the error bars
--nobin Don't use the binned errorbars
--nolegend Turn off the legend
--ttest Perform a ttest
--hline=<val> Include a horizontal line at <val> in the averaged plot
--hlabel=<hl> A legend label for the horizontal line.
--title=<title> A title for the plots.
--savefig=<figure> A filename for saved plots (extensions supported by active matplotlib backend).
--quiet Suppress output
usage: reduce-one.py [-h] [-T T] [-N N] [-n N] [-t TAU] [-u MU] [-L L] [-V V] -r {T,N,n,u,t,L,V,M} [--canonical] [-R R] [-s SKIP] [-e ESTIMATOR] [-i PIMCID] [base_dir]
Reduce quantum Monte Carlo output over some parameter.
positional arguments:
base_dir The base directory where the data files to be reduced are located.
optional arguments:
-h, --help show this help message and exit
-T T, --temperature T
simulation temperature in Kelvin
-N N, --number-particles N
number of particles
-n N, --density N number density in Angstroms^{-d}
-t TAU, --imag-time-step TAU
imaginary time step
-u MU, --chemical-potential MU
chemical potential in Kelvin
-L L, --Lz L Length in Angstroms
-V V, --volume V volume in Angstroms^d
-r {T,N,n,u,t,L,V,M}, --reduce {T,N,n,u,t,L,V,M}
variable name for reduction [T,N,n,u,t,L,V,W,M]
--canonical are we in the canonical ensemble?
-R R, --radius R radius in Angstroms
-s SKIP, --skip SKIP number of measurements to skip [0]
-e ESTIMATOR, --estimator ESTIMATOR
specify a single estimator to reduce
-i PIMCID, --pimcid PIMCID
specify a single pimcid