Getting Started

Brief overview of pymchelper and how to install it.

Introduction

Let us assume we are running particle transport simulation using Fluka MC code. Two estimators: particle fluence and deposited energy are defined in the input file. To get high statistics run is parallelised into 100 different jobs, each of them producing 2 binary files. That gives 200 binary output files, typically following *_fort.* pattern.

pymchelper comes with convertmc program which simplifies postprocessing of such binary data.

To convert all binary files into two text files (one for energy and one for particle fluence scorer) type

convertmc txt --many "*_fort.*"

pymchelper will automatically figure out how many scorers were defined and which files to merge. Two new files will be created: 21.txt and 22.txt which can further processed. By default they will contain 5 columns: X,Y,Z coordinates, data and error column:

0.0000000E+00  0.0000000E+00  0.5000000E-02  0.1881775795482099E-02  0.2157818940293169E-04
0.0000000E+00  0.0000000E+00  0.1500000E-01  0.1893831696361303E-02  0.2251415707395123E-04
0.0000000E+00  0.0000000E+00  0.2500000E-01  0.1887728041037917E-02  0.1936414681456153E-04
0.0000000E+00  0.0000000E+00  0.3500000E-01  0.1897429465316236E-02  0.1944586725074595E-04

In case estimated value was scored on 1-dimensional grid, it can be easily plotted. Same thing might be done if 2-D scoring grid was used - then a heatmap type plot can be produced. In order to get such plots instead of text files, replace first argument with image:

convertmc image --many "*_fort*"

Two new PNG files will be created which can be directly opened, for example

sample file 21.txt.png generated with image converter

If you are SHIELD-HIT12A user, same effect can be achieved by processing *.bdo binary files.

These are just basic examples. To learn more about additional features, proceed to User’s Guide. Among these features are:

  • reading binary *.bdo files generated by SHIELD-HIT12A code
  • reading binary *_fort* files generated by FLUKA
  • calculating standard error when averaging many files
  • writing PNG images (1 and 2D plots)
  • writing tabulated text files
  • writing gnuplot scripts

Quick Installation Guide

Be sure to have Python framework installed, then type:

pip install pymchelper

In case you don’t have administrator rights, add --user flag to pip command. In this situation converter will be probably installed in ~/.local/bin directory.

License

pymchelper is licensed under MIT LICENCE.