Error running grid MC in Zr-O demo (TypeError: cannot unpack non-iterable NoneType object)

[sjtuzhanglei]

Dear CASM developers,

When running the demo MC folder 'grid' for ZrO, I got errors:

Traceback (most recent call last): File "/global/u2/v/voz5005/ZrO/mc/grid/run.py", line 204, in submit_T_up(T_grid, xi_grid) File "/global/u2/v/voz5005/ZrO/mc/grid/run.py", line 105, in submit_T_up monte_cmd(tinput) File "/global/u2/v/voz5005/ZrO/mc/grid/run.py", line 56, in monte_cmd stdout, stderr, returncode = proj.command('monte -s input.json') TypeError: cannot unpack non-iterable NoneType object

Input file is:

import subprocess import json import os import numpy as np import copy from casm.project import Project import sh as pbs

print (""" Running a grid of Monte Carlo calculations (This may take a while)

""")

Grid space to run Monte Carlo calculations (min, max, delta)

T_grid = (200, 1000, 100) xi_grid = (-2.2, -1.6, 0.05)

d = { "mode" : "incremental", "dependent_runs" : True, "motif" : { "configname" : "restricted_auto", }, "initial_conditions" : { "param_chem_pot" : { "a" : -3.00 }, "temperature" : 1000.0, "tolerance" : 0.01 }, "final_conditions" : { "param_chem_pot" : { "a" : -3.00 }, "temperature" : 200.0, "tolerance" : 0.01 }, "incremental_conditions" : { "param_chem_pot" : { "a" : 0.0 }, "temperature" : -100.0, "tolerance" : 0.01 } }

proj = None

For submission on personal computer...

def monte_cmd(input): global proj if proj is None: proj = Project() stdout, stderr, returncode = proj.command('monte -s input.json') print (stdout)

For submission on PRISMS cluster...

#def monte_cmd(input):

jobname = os.path.basename(os.getcwd())

cmd = """

#source ~/privatemodules/casm #casm monte -s input.json #"""

job = pbs.PrismsDebugJob(name = jobname,

nodes = "1",

ppn = "1",

walltime = "6:00:00",

pmem = "3800mb",

exetime = None,

message = "a",

email = None,

command = cmd,

auto = False).submit()

def submit_T_up(T_grid, xi_grid): index = 0 for xi in np.arange(*xi_grid): path = "T_up." + str(index)

tinput = copy.deepcopy(input)
tinput['driver'] = d
tinput['driver']['initial_conditions']['temperature'] = T_grid[0]
tinput['driver']['final_conditions']['temperature'] = T_grid[1]
tinput['driver']['incremental_conditions']['temperature'] = T_grid[2]

tinput['driver']['initial_conditions']['param_chem_pot']['a'] = xi
tinput['driver']['final_conditions']['param_chem_pot']['a'] = xi
tinput['driver']['incremental_conditions']['param_chem_pot']['a'] = 0.0

cwd = os.getcwd()
if not os.path.exists(path):
  os.mkdir(path)
os.chdir(path)
print ("Running:", path)

with open('input.json','w') as f:
  json.dump(tinput, f)

monte_cmd(tinput)

os.chdir(cwd)
index += 1

def submit_T_down(T_grid, xi_grid): index = 0 for xi in np.arange(*xi_grid): path = "T_down." + str(index)

tinput = copy.deepcopy(input)
tinput['driver'] = d
tinput['driver']['initial_conditions']['temperature'] = T_grid[1]
tinput['driver']['final_conditions']['temperature'] = T_grid[0]
tinput['driver']['incremental_conditions']['temperature'] = -T_grid[2]

tinput['driver']['initial_conditions']['param_chem_pot']['a'] = xi
tinput['driver']['final_conditions']['param_chem_pot']['a'] = xi
tinput['driver']['incremental_conditions']['param_chem_pot']['a'] = 0.0

cwd = os.getcwd()
if not os.path.exists(path):
  os.mkdir(path)
os.chdir(path)
print ("Running:", path)

with open('input.json','w') as f:
  json.dump(tinput, f)

monte_cmd(tinput)

os.chdir(cwd)
index += 1

def submit_xi_up(T_grid, xi_grid): index = 0 for T in np.arange(*T_grid): path = "xi_up." + str(index)

tinput = copy.deepcopy(input)
tinput['driver'] = d
tinput['driver']['initial_conditions']['temperature'] = T
tinput['driver']['final_conditions']['temperature'] = T
tinput['driver']['incremental_conditions']['temperature'] = 0.0

tinput['driver']['initial_conditions']['param_chem_pot']['a'] = xi_grid[0]
tinput['driver']['final_conditions']['param_chem_pot']['a'] = xi_grid[1]
tinput['driver']['incremental_conditions']['param_chem_pot']['a'] = xi_grid[2]

cwd = os.getcwd()
if not os.path.exists(path):
  os.mkdir(path)
os.chdir(path)
print ("Running:", path)

with open('input.json','w') as f:
  json.dump(tinput, f)

monte_cmd(tinput)

os.chdir(cwd)
index += 1

def submit_xi_down(T_grid, xi_grid): index = 0 for T in np.arange(*T_grid): path = "xi_down." + str(index)

tinput = copy.deepcopy(input)
tinput['driver'] = d
tinput['driver']['initial_conditions']['temperature'] = T
tinput['driver']['final_conditions']['temperature'] = T
tinput['driver']['incremental_conditions']['temperature'] = 0.0

tinput['driver']['initial_conditions']['param_chem_pot']['a'] = xi_grid[1]
tinput['driver']['final_conditions']['param_chem_pot']['a'] = xi_grid[0]
tinput['driver']['incremental_conditions']['param_chem_pot']['a'] = -xi_grid[2]

cwd = os.getcwd()
if not os.path.exists(path):
  os.mkdir(path)
os.chdir(path)
print ("Running:", path)

with open('input.json','w') as f:
  json.dump(tinput, f)

monte_cmd(tinput)

os.chdir(cwd)
index += 1

with open('metropolis_grand_canonical.json','r') as f: input = json.load(f)

submit_T_up(T_grid, xi_grid) submit_T_down(T_grid, xi_grid) submit_xi_up(T_grid, xi_grid) submit_xi_down(T_grid, xi_grid)

[bpuchala]

Thanks for the report on this, I'll take a look when I can. In the meantime, it can still serve as a guide to how to grid up thermodynamic space for free energy integration.