Homec4science
Diffusion Diffusion limited mixed aggregation (master) Browse

dlma/scripts/scaling_vs_agg_tolerance/setup_simulation_dlca.py
fba57248c1b0master

Last Change
Raw File

setup_simulation_dlca.py
View Options

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri Jun 4 13:56:09 2021
@author: samarth
"""
import pandas as pd
import os
import sys
from pathlib import Path
from joblib import Parallel, delayed
from matplotlib import pyplot as plt
import numpy as np
from sklearn.linear_model import LinearRegression
from random import random, seed
import subprocess
def create_dirs(phi, agg_distance):
global D
D = dim
# global params_dir
phi_str = format(phi, '.5f')
agg_str = format(agg_distance, '.3f')
params_dir = './scaling/params/phi='+phi_str+'/'+agg_str+'/'
path = Path(params_dir)
try:
path.mkdir(parents=True, exist_ok=True)
except OSError:
pass
else:
pass
# global results_dir
results_dir = './scaling/results/phi='+phi_str+'/'+agg_str+'/'
path = Path(results_dir)
try:
path.mkdir(parents=True, exist_ok=True)
except OSError:
pass
else:
pass
# global config_files_dir
config_files_dir = './scaling/config_files/phi='+phi_str+'/'+agg_str+'/'
path = Path(config_files_dir)
try:
path.mkdir(parents=True, exist_ok=True)
except OSError:
pass
else:
pass
def run_simulation(phi, agg_distance, rng_seed):
phi_str = format(phi, '.5f')
agg_str = format(agg_distance, '.3f')
params_dir = './scaling/params/phi='+phi_str+'/'+agg_str+'/'
results_dir = './scaling/results/phi='+phi_str+'/'+agg_str+'/'
config_files_dir = './scaling/config_files/phi='+phi_str+'/'+agg_str+'/'
lattice=0
alpha=0.5
seedMass=1e12
N=25000
params_file = params_dir+'params_'+str(rng_seed)+'.csv'
file = open(params_file, "w")
file.write("system=dlma\n")
file.write("lattice="+str(lattice)+"\n")
file.write("aggregation_type=normal\n")
file.write("aggregation_condition=mass\n")
file.write("bind=normal\n")
file.write("movement=brownian\n")
file.write("agg_dist_tolerance="+format(agg_distance,'.2f')+"\n")
file.write("D="+str(D)+"\n")
file.write("seed_pct=100\n")
for axis in range(D):
file.write("x"+str(axis)+"_bc=periodic\n")
file.write("N="+str(N)+"\n")
file.write("phi="+str(phi)+"\n")
file.write("alpha="+str(alpha)+"\n")
file.write("seedMass="+str(seedMass)+"\n")
file.write("rng_seed="+str(rng_seed)+"\n")
file.close()
print("phi = {}, agg_dist = {}, rng_seed={}, simulation".format(phi,agg_distance,rng_seed))
config_filename=config_files_dir+str(rng_seed)+'.csv'
command = '../../simulation_files/bin/simulation '+params_file+' '+config_filename
os.system(command)
results_filename=results_dir+'lb_'+str(rng_seed)+'.csv'
command="../../postprocessing/bin/lb_bonds_clusterwise_invA "+config_filename+" "+ results_filename
os.system(command)
results_filename=results_dir+'sq_'+str(rng_seed)+'.csv'
command = "../../postprocessing/bin/long_range_scattering "+config_filename+" 1000 0.01 50 2000 "+ results_filename
os.system(command)
os.system('cd ../../simulation_files; make simulation')
os.system('cd ../../postprocessing/; make all')
seed(1)
seeds=100
seed_pct_range=[100]
dim=3
agg_dist_tolerance_range = [0.15, 0.2, 0.25, 0.175, 0.225]
phi_min = 0.1
phi_max = 0.3
num_i = 10
delta_phi = np.log(phi_max/phi_min)/(1.*num_i)
phi_range = []
for i in range(num_i):
phi_temp = phi_min * np.exp(1.*i*delta_phi)
phi_range.append(phi_temp)
all_params = []
for phi in phi_range:
for agg_dist_tolerance in agg_dist_tolerance_range:
all_params.append([phi, agg_dist_tolerance])
for param in all_params:
create_dirs(param[0], param[1])
all_params.clear()
agg_dist_tolerance_range_short = [0.15, 0.2, 0.25]
# 50 x 3 * 10 = 1500
for phi in phi_range:
for agg_dist_tolerance in agg_dist_tolerance_range_short:
for i in range(50, seeds):
all_params.append([phi, agg_dist_tolerance, i])
agg_dist_tolerance_range_ext = [0.175, 0.225]
for phi in phi_range:
for agg_dist_tolerance in agg_dist_tolerance_range_ext:
for i in range(seeds):
all_params.append([phi, agg_dist_tolerance, i])
for i in range(50):
all_params.append([phi_range[-1], agg_dist_tolerance_range_short[-1], i])
num_processors = 80
Parallel(n_jobs=num_processors)(delayed(run_simulation)(param[0], param[1], param[2]) for param in all_params)
os.system('cd ../../simulation_files/; make clean')
os.system('cd ../../postprocessing/; make clean')
c4science · Help