Make your own custom Action

In this example, a user can follow the instructions in /feasst/plugin/example/include/action_example.h to make their own custom Action.

In this relatively trivial example, the newly implemented Action forces an Analyze or Modify to write to file.

import sys
import subprocess
import numpy as np
import pandas as pd

def run_fst():
    with open("launch3.txt", "w") as myfile: myfile.write("""
MonteCarlo
Configuration particle_type0 /feasst/particle/atom.fstprt cubic_side_length 8
Potential Model IdealGas
ThermoParams beta 1 chemical_potential0 1
FlatHistogram Macrostate MacrostateNumParticles width 1 max 4 min 0 Bias TransitionMatrix min_sweeps 2
TrialTransfer particle_type 0
CriteriaUpdater trials_per_update 1e2
CriteriaWriter trials_per_write 1e2 output_file action_ex_crit.csv
Energy output_file en.csv trials_per_write 1e2 multistate true
PairDistribution output_file grig.csv trials_per_write 1e2
Run until complete
ActionExample analyze_name CriteriaWriter
ActionExample analyze_name Energy
ActionExample modify_name PairDistribution
""")
    syscode = subprocess.call("../../../build/bin/fst < launch3.txt", shell=True, executable='/bin/bash')
    if syscode > 0: sys.exit(1)

Check that, with “ActionExample analyze_name CriteriaWriter”, the number of sweeps in the writen file is equal to the number of sweeps requested (i.e., 2).

run_fst()
file1 = open('action_ex_crit.csv', 'r')
lines = file1.readlines()
file1.close()
exec('iprm={' + lines[0][1:]+'}', globals())
print(iprm)
assert iprm['num_sweeps'] == 2

# Usage: ./fst < file.txt
FEASST version 0.25.6
MonteCarlo
Configuration cubic_side_length 8 particle_type0 /home/user/feasst/particle/atom.fstprt
Potential Model IdealGas
ThermoParams beta 1 chemical_potential0 1
FlatHistogram Bias TransitionMatrix Macrostate MacrostateNumParticles max 4 min 0 min_sweeps 2 width 1
TrialTransfer particle_type 0
CriteriaUpdater trials_per_update 1e2
CriteriaWriter output_file action_ex_crit.csv trials_per_write 1e2
Energy multistate true output_file en.csv trials_per_write 1e2
PairDistribution output_file grig.csv trials_per_write 1e2
Run until complete
# initializing random number generator with seed: 1734454620
ActionExample analyze_name CriteriaWriter
ActionExample analyze_name Energy
ActionExample modify_name PairDistribution
{'num_sweeps': 2, 'soft_min': 0, 'soft_max': 4}

Did this tutorial work as expected? Did you find any inconsistencies or have any comments? Please contact us. Any feedback is appreciated!