def run(args):
"""Runs the matdb setup and cleanup to produce database files.
"""
if args is None:
return
cmd = args["cmd"]
print(cmd)
keyword = cmd[0].strip()
print(keyword)
if keyword == "calc-grade":
touch("state.mvs")
touch("temp1.cfg")
elif keyword == "select-add":
template_root = path.join(_get_reporoot(), "tests", "fitting", "files")
touch("new_training.cfg")
# src = path.join(template_root,"mtp_new_training.cfg")
# dest = "new_training.cfg"
# copyonce(src, dest)
elif keyword == "convert-cfg":
template_root = path.join(_get_reporoot(), "tests", "mtp", "training")
for i in range(1, 11):
src = path.join(template_root, "POSCAR{0}".format(i))
dest = "POSCAR{0}".format(i)
copyonce(src, dest)
def Act(tmpdir):
target = relpath("./tests/AgPd/matdb.yml")
dbdir = str(tmpdir.join("active_db"))
mkdir(dbdir)
copyonce(target, path.join(dbdir, "matdb.yml"))
target = path.join(dbdir, "matdb")
seed_root = path.join(dbdir, "seed")
if not path.isdir(seed_root):
mkdir(seed_root)
for i in range(1, 4):
cfg_target = path.join(seed_root, "Pd{0}".format(i))
cfg_source = path.join(_get_reporoot(), "tests", "database", "files",
"Pd", "POSCAR{0}".format(i))
copyonce(cfg_source, cfg_target)
cntrl = Controller(target, dbdir)
db = Database("active", dbdir, cntrl, [{"type": "active.Active"}], {}, 0)
tcntrl = TController(db=db, root=dbdir, fits={})
dbargs = {"root": dbdir, "parent": db, "calculator": tcntrl.db.calculator}
result = Active(**dbargs)
return result
def test_prot_to_cfg():
"""Tests the conversion of a prototype structure to a cfg file.
"""
from matdb.fitting.mtp import _prot_to_cfg
from matdb.utility import _get_reporoot, touch, chdir
from os import path, remove
import tarfile
template_root = path.join(_get_reporoot(), "matdb", "templates")
if not path.isdir(path.join(template_root, "uniqueUnaries")):
with chdir(template_root):
tarf = "prototypes.tar.gz"
tar = tarfile.open(tarf, "r:gz")
tar.extractall()
tar.close()
source = path.join(template_root, "uniqueUnaries", "A10_POSCAR.orig")
species = ["Al"]
relax_file = "relax.cfg"
type_map = {0:0}
root = "."
min_atoms = 1
max_atoms = None
touch(relax_file)
_prot_to_cfg(source, species, relax_file, type_map, root, min_atoms, max_atoms)
assert path.isfile(relax_file)
lc = 0
with open(relax_file, "r") as f:
for line in f:
lc += 1
assert lc == 12
source = path.join(template_root, "uniqueBinaries", "b31_POSCAR.orig")
species = ["Al", "Cu"]
relax_file = "relax.cfg"
type_map = {0:0, 1:1}
root = "."
min_atoms = 1
max_atoms = 4
_prot_to_cfg(source, species, relax_file, type_map, root, min_atoms, max_atoms)
lc = 0
with open(relax_file, "r") as f:
for line in f:
lc += 1
assert lc == 12
remove(relax_file)
def Pd(tmpdir):
target = relpath("./tests/Pd/matdb.yml")
dbdir = str(tmpdir.join("manual_db"))
mkdir(dbdir)
copyonce(target, path.join(dbdir, "matdb.yml"))
seed_root = path.join(dbdir, "seed")
if not path.isdir(seed_root):
mkdir(seed_root)
for i in range(1, 4):
cfg_target = path.join(seed_root, "Pd{0}".format(i))
cfg_source = path.join(_get_reporoot(), "tests", "database", "files",
"Pd", "POSCAR{0}".format(i))
copyonce(cfg_source, cfg_target)
target = path.join(dbdir, "matdb")
cntrl = Controller(target, dbdir)
return cntrl
def Pd_not_extractable(tmpdir):
target = relpath("./tests/Pd/matdb_not_extractable.yml")
dbdir = str(tmpdir.join("manual_not_extractale_db"))
mkdir(dbdir)
copyonce(target, path.join(dbdir, "matdb.yml"))
seed_root = path.join(dbdir, "seed")
if not path.isdir(seed_root):
mkdir(seed_root)
cfg_target = path.join(seed_root, "Pd")
cfg_source = path.join(_get_reporoot(), "tests", "database", "files", "Pd",
"POSCAR1")
copyonce(cfg_source, cfg_target)
target = path.join(dbdir, "matdb")
cntrl = Controller(target, dbdir)
# these 3 lines are just added to cover dbargs["calculator"] not None in __init__
db = Database("phonon", dbdir, cntrl, [{"type": "simple.Manual"}], {}, 0)
dbargs = {"root": dbdir, "parent": db, "calculator": cntrl.calculator}
mdb = Manual(extractable=False, **dbargs)
return cntrl
def create_to_relax(setup_args):
"""Creates the to-relax.cfg file from the passed in args dictionary.
Args:
setup_args (dict): A dictionary containing the arguments needed to
construct the potential.
"""
args = setup_args["phenum_args"]
species = np.unique(setup_args["species"])
crystals = setup_args["crystals"]
min_atoms = setup_args["min_atoms"]
max_atoms = setup_args["max_atoms"]
root = setup_args["root"]
touch(args["outfile"])
for crystal in crystals:
prot_map = {
0: "uniqueUnaries",
1: "uniqueBinaries",
2: "uniqueTernaries"
}
if crystal == "prototypes":
# The prototypes are saved into the file prototypes.tar.gz, if
# this is the first time prototypes has been run we need to unpack it.
template_root = path.join(_get_reporoot(), "matdb", "templates")
if not path.isdir(path.join(template_root, "uniqueUnaries")):
with chdir(template_root):
tarf = "prototypes.tar.gz"
tar = tarfile.open(tarf, "r:gz")
tar.extractall()
tar.close()
for size in range(len(species)):
cand_path = path.join(template_root, prot_map[size])
structures = glob("{0}/*".format(cand_path))
perms = [list(i) for i in permutations(species, r=size + 1)]
for fpath, perm in product(structures, perms):
type_map = {}
for i, s in enumerate(species):
if s in perm:
type_map[perm.index(s)] = i
_prot_to_cfg(fpath, perm, args["outfile"], type_map, root,
min_atoms, max_atoms)
else:
# eventually we'll want to replace this with an actual
# enumeration over the options the user specifies.
infile = path.join(
_get_reporoot(), "matdb", "templates",
"struct_enum.out_{0}_{1}".format(len(species), crystal))
expected_min_atoms = 1
if crystal == "hcp":
min_atoms = 2
expected_min_atoms = 2
if min_atoms != expected_min_atoms or max_atoms is not None:
with open(infile, "r") as f:
min_num = None
max_num = None
last_num = 1
past_start = False
for line in f:
if line.split()[0] == "start":
past_start = True
continue
if past_start:
data = line.strip().split()
lab = data[-2]
last_num = int(data[0])
if len(lab) == min_atoms and min_num is None:
min_num = int(data[0])
if len(lab) > max_atoms and max_num is None:
max_num = int(data[0])
break
# In case in the input file, there is no configuration
# that has the number of atoms bigger than max_atoms,
# the max_num will be set to the number in the last
# line.
if max_num is None:
max_num = last_num
args["structures"] = range(min_num, max_num)
args["input"] = infile
_make_structures(args)
def __init__(self,
name="prototype",
structures=None,
ran_seed=None,
permutations=None,
**dbargs):
self.name = name
self.seeded = False
dbargs["prefix"] = "P"
dbargs["cls"] = Prototypes
if "Prototypes" not in dbargs['root']:
from os import mkdir
new_root = path.join(dbargs['root'], "Prototypes")
if not path.isdir(new_root):
mkdir(new_root)
dbargs['root'] = new_root
super(Prototypes, self).__init__(**dbargs)
self.in_structures = structures
self.ran_seed = ran_seed
self.permutations = permutations
self.species = self.database.parent.species
#Make sure that we override the global calculator default values with
#those settings that we know are needed for good phonon calculations.
calcargs = self.database.calculator.copy()
if "calculator" in dbargs:
if dbargs["calculator"] is not None and "name" in dbargs[
"calculator"]:
calcargs.update(dbargs["calculator"])
dbargs["calculator"] = calcargs
# The prototypes are saved into the file prototypes.tar.gz, if
# this is the first time prototypes has been run we need to unpack it.
template_root = path.join(_get_reporoot(), "matdb", "templates")
if not path.isdir(path.join(template_root, "uniqueUnaries")):
import tarfile
with chdir(template_root):
tarf = "prototypes.tar.gz"
tar = tarfile.open(tarf, "r:gz")
tar.extractall()
tar.close()
# parse the structures to make a list of paths to the source folders for the
if self.ran_seed is not None:
import random
random.seed(self.ran_seed)
self.puuids = None
self._load_puuids()
self.nconfigs = 0
self.structures = {}
for k, v in structures.items():
if k.lower() == "unary":
cand_path = path.join(template_root, "uniqueUnaries")
elif k.lower() == "binary":
cand_path = path.join(template_root, "uniqueBinaries")
elif k.lower() == "ternary":
cand_path = path.join(template_root, "uniqueTernaries")
else: # pragma: no cover
msg.warn(
"Must specify the system size, i.e., unary, binary, or "
"ternary. {} not recognized".format(k))
continue
if isinstance(v, list):
self.structures[k.lower()] = []
for prot in v:
files = glob("{0}/*{1}*".format(cand_path, prot))
if len(files) < 1: # pragma: no cover
msg.warn(
"No prototypes of size {0} matched the string "
"{1}".format(k, prot))
else:
self.structures[k.lower()].extend(files)
elif isinstance(v, str) and v == "all":
files = glob("{0}/*".format(cand_path))
self.structures[k.lower()] = files
elif isinstance(v, int):
from random import shuffle
files = glob("{0}/*".format(cand_path))
shuffle(files)
keep = files[:v]
self.structures[k.lower()] = keep
else: #pragma: no cover
msg.err(
"Couldn't parse {0} structures for {1} case. Must be either "
"a list of file names, 'all', or an int.".format(v, k))
if self.permutations is not None and k.lower(
) in self.permutations.keys():
self.nconfigs += len(self.structures[k.lower()]) * len(
self.permutations[k.lower()])
else:
if k.lower() == "unary":
self.nconfigs += len(self.structures[k.lower()]) * 3
elif k.lower() == "binary" or k.lower() == "ternary":
self.nconfigs += len(self.structures[k.lower()]) * 6
else: #pragma: no cover
continue
def test_create_to_relax():
"""Tests the creation of the to_relax_file.
"""
from matdb.fitting.mtp import create_to_relax
from matdb.utility import _get_reporoot
from os import remove, path
import shutil
template_root = path.join(_get_reporoot(), "matdb", "templates")
if path.isdir(path.join(template_root, "uniqueUnaries")):
shutil.rmtree(path.join(template_root, "uniqueUnaries"), ignore_errors=True)
shutil.rmtree(path.join(template_root, "uniqueBinaries"), ignore_errors=True)
shutil.rmtree(path.join(template_root, "uniqueTernaries"), ignore_errors=True)
setup_args = {}
target = "to-relax.cfg"
if path.isfile(target):
remove(target)
setup_args["phenum_args"] = {"config":"t", "species": ["Al", "Cu"], "structures": "all",
"debug": False, "example": False, "displace":0.0,
"mink":True, "outfile":target, "verbose": None,
"rattle":0.0, "mapping": None}
setup_args["species"] = ["Al", "Cu"]
setup_args["crystals"] = ["fcc"]
setup_args["min_atoms"] = 1
setup_args["max_atoms"] = 2
setup_args["root"] = "."
create_to_relax(setup_args)
struct_count = 0
with open(target,"r") as f:
for line in f:
if "BEGIN_CFG" in line:
struct_count += 1
assert struct_count == 4
remove(target)
setup_args["crystals"] = ["fcc", "prototypes", "hcp"]
create_to_relax(setup_args)
struct_count = 0
with open(target,"r") as f:
for line in f:
if "BEGIN_CFG" in line:
struct_count += 1
assert struct_count == 141
remove(target)
setup_args["crystals"] = ["fcc"]
setup_args["min_atoms"] = 12
setup_args["max_atoms"] = 14
create_to_relax(setup_args)
struct_count = 0
with open(target,"r") as f:
for line in f:
if "BEGIN_CFG" in line:
struct_count += 1
assert struct_count == 7139
remove(target)
def test_command_functions(mtpdb):
"""All the commands are iterative, so we'll have to test them one
after the other.
"""
#first build the database directory and the initial database of
#structures.
from matdb.utility import copyonce, _get_reporoot, touch
from os import path, mkdir, remove, rename
from matdb.atoms import Atoms
seed_root = path.join(mtpdb.root, "seed")
mkdir(seed_root)
templates = path.join(_get_reporoot(), "tests", "mtp", "training")
for i in range(1,11):
target = path.join(seed_root, "vasp.{0}".format(i))
source = path.join(templates, "POSCAR{0}".format(i))
copyonce(source, target)
mtpfit = mtpdb.trainers.fits['CoWV_mtp'].sequences['CoWV_mtp'].steps['mtp']
cntrl_root = mtpdb.root
db_root = path.join(cntrl_root, "Manual", "test.manual")
mtpdb.setup()
files = ["OUTCAR{}", "CONTCAR{}"]
for i in range(1,11):
target = path.join(db_root, "vasp.{0}".format(i), "S1.1")
for f in files:
copyonce(path.join(templates, f.format(i)), path.join(target, f.format('')))
mtpdb.extract()
touch(path.join(mtpfit.root, "new_training.cfg"))
touch(path.join(mtpfit.root, "relaxed.cfg"))
# With the initial database setup we can now run the mtp commands.
# First test the initial training setup script
cmd_template = mtpfit.command()
target = path.join(mtpfit.root, "status.txt")
bc = path.join(mtpfit.root, "status.txt_train")
copyonce(target,bc)
assert path.isfile(target)
with open(target, "r") as f:
line = f.read()
assert line.strip() == "relax_setup 1 0"
assert path.isfile(path.join(mtpfit.root, "relax.ini"))
assert path.isfile(path.join(mtpfit.root, "pot.mtp"))
assert cmd_template == mtpfit._train_template()
# Now we setup for the test of the relaxation_setup
copyonce(path.join(mtpfit.root, "pot.mtp"), path.join(mtpfit.root, "Trained.mtp_"))
cmd_template = mtpfit.command()
assert "matdb_mtp_to_relax.py" in cmd_template
with open(target, "r") as f:
line = f.read()
assert line.strip() == "relax 1 0"
rename(bc, target)
copyonce(target,bc)
copyonce(path.join(mtpfit.root, "pot.mtp"), path.join(mtpfit.root, "Trained.mtp_"))
touch(path.join(mtpfit.root, "unrelaxed.cfg"))
mtpfit.use_unrelaxed = True
cmd_template = mtpfit.command()
assert cmd_template == mtpfit._relax_template()
with open(target, "r") as f:
line = f.read()
assert line.strip() == "select 1 0"
mtpfit.use_unrelaxed = False
rename(bc, target)
copyonce(path.join(mtpfit.root, "pot.mtp"), path.join(mtpfit.root, "Trained.mtp_"))
touch(path.join(mtpfit.root, "candidate.cfg"))
cmd_template = mtpfit.command()
assert cmd_template == mtpfit._relax_template()
with open(target, "r") as f:
line = f.read()
assert line.strip() == "select 1 0"
# Now we can test the select step
for i in range(72):
touch(path.join(mtpfit.root, "candidate.cfg_{0}".format(i)))
cmd_template = mtpfit.command()
assert path.isfile(path.join(mtpfit.root, "candidate.cfg"))
with open(target, "r") as f:
line = f.read()
assert line.strip() == "add 1 0"
def test_train_setup(mtpdb):
"""Tests the mtp training setup.
"""
#test make_train_cfg
#first build the database directory.
from matdb.utility import copyonce, _get_reporoot
from os import path, mkdir, remove
from matdb.atoms import Atoms
seed_root = path.join(mtpdb.root, "seed")
mkdir(seed_root)
templates = path.join(_get_reporoot(), "tests", "mtp", "training")
for i in range(1,11):
target = path.join(seed_root, "vasp.{0}".format(i))
source = path.join(templates, "POSCAR{0}".format(i))
copyonce(source, target)
mtpfit = mtpdb.trainers.fits['CoWV_mtp'].sequences['CoWV_mtp'].steps['mtp']
cntrl_root = mtpdb.root
db_root = path.join(cntrl_root, "Manual", "test.manual")
mtpdb.setup()
files = ["OUTCAR{}", "CONTCAR{}"]
for i in range(1,11):
target = path.join(db_root, "vasp.{0}".format(i), "S1.1")
for f in files:
copyonce(path.join(templates, f.format(i)), path.join(target, f.format('')))
mtpdb.extract()
mtpfit._make_train_cfg(1)
assert path.isfile(path.join(mtpfit.root, "train.cfg"))
struct_count = 0
with open(path.join(mtpfit.root, "train.cfg"),"r") as f:
for line in f:
if "BEGIN_CFG" in line:
struct_count += 1
assert struct_count == 10
new_configs = []
for i in range(1,11):
source = path.join(templates, "POSCAR{0}".format(i))
new_configs.append(Atoms(source))
mtpfit.active.add_configs(new_configs, 2)
mtpfit.active.setup()
act_root = path.join(mtpdb.root, "Active", "active.CoWV_mtp")
files = ["OUTCAR{}", "CONTCAR{}"]
for i in range(1,11):
target = path.join(act_root, "Ac.{0}".format(i))
for f in files:
copyonce(path.join(templates, f.format(i)), path.join(target, f.format('')))
# This is trying to manipulate an NOT extractable structure
target = path.join(act_root, "Ac.{0}".format(3), "OUTCAR")
remove(target)
source = path.join(_get_reporoot(), "tests", "files", "VASP", "OUTCAR_incomplete")
copyonce(source, target)
mtpfit.active.last_iteration = None
mtpfit._make_train_cfg(2)
assert path.isfile(path.join(mtpfit.root, "train.cfg"))
struct_count = 0
with open(path.join(mtpfit.root, "train.cfg"),"r") as f:
for line in f:
if "BEGIN_CFG" in line:
struct_count += 1
assert struct_count == 19
mtpfit.active.last_iteration = None
remove(mtpfit.active.iter_file)
with pytest.raises(IOError):
mtpfit._make_train_cfg(2)