def main():
setup_timestamp_logging()
# Load in the force field
force_field_path = "smirnoff99Frosst-1.1.0.offxml"
force_field_source = SmirnoffForceFieldSource.from_path(force_field_path)
# Load in the data set containing the pure and binary properties.
data_set = PhysicalPropertyDataSet.from_json("pure_data_set.json")
data_set.merge(PhysicalPropertyDataSet.from_json("binary_data_set.json"))
# Set up a server object to run the calculations using.
server = setup_server(backend_type=BackendType.LocalGPU,
max_number_of_workers=1,
port=8001)
with server:
# Request the estimates.
property_estimator = EvaluatorClient(
ConnectionOptions(server_port=8001))
for calculation_layer in ["SimulationLayer", "ReweightingLayer"]:
options = RequestOptions()
options.calculation_layers = [calculation_layer]
parameter_gradient_keys = [
ParameterGradientKey(tag="vdW",
smirks="[#6X4:1]",
attribute="epsilon"),
ParameterGradientKey(tag="vdW",
smirks="[#6X4:1]",
attribute="rmin_half"),
]
request, _ = property_estimator.request_estimate(
property_set=data_set,
force_field_source=force_field_source,
options=options,
parameter_gradient_keys=parameter_gradient_keys,
)
# Wait for the results.
results, _ = request.results(True, 5)
layer_name = re.sub(r"(?<!^)(?=[A-Z])", "_",
calculation_layer).lower()
results.json(f"pure_binary_{layer_name}.json", True)
def main(input_data_set_path, server_port):
# Create the options which propertyestimator should use.
estimator_options = RequestOptions()
# Choose which calculation layers to make available.
estimator_options.calculation_layers = ["SimulationLayer"]
# Load in the training data set and create schemas for each of the types
# of property to be calculated.
training_set = PhysicalPropertyDataSet.from_json(input_data_set_path)
# Zero out any undefined uncertainties due to a bug in ForceBalance.
for physical_property in training_set:
physical_property.uncertainty = 0.0 * physical_property.default_unit()
data_set_path = "training_set.json"
training_set.json(data_set_path, format=True)
# Create the force balance options
target_options = Evaluator_SMIRNOFF.OptionsFile()
target_options.connection_options = ConnectionOptions(
server_address="localhost", server_port=server_port)
target_options.estimation_options = estimator_options
target_options.data_set_path = data_set_path
# Set the property weights and denominators.
target_options.weights = {x: 1.0 for x in training_set.property_types}
target_options.denominators = calculate_denominators(training_set)
# Save the options to file.
with open("options.json", "w") as file:
file.write(target_options.to_json())
def find_training_smiles():
"""Returns the smiles of all of the substances which
appeared in the training set.
Returns
-------
list of tuple of str
The smiles patterns of the training substances.
"""
# Find those alcohols which were included in the training set
training_set = PhysicalPropertyDataSet.from_json(
os.path.join(
"..",
"..",
"..",
"pure_mixture_optimisation",
"force_balance",
"alcohol_ester",
"h_mix_rho_x_rho_pure_h_vap",
"targets",
"mixture_data",
"training_set.json",
)
).to_pandas()
training_smiles = data_frame_to_smiles_tuples(training_set)
training_smiles = set(x for y in training_smiles for x in y)
return training_smiles
def main():
setup_timestamp_logging()
# Load in the force field
force_field_path = "openff-1.0.0-refit.offxml"
force_field_source = SmirnoffForceFieldSource.from_path(force_field_path)
# Load in the test set.
data_set = PhysicalPropertyDataSet.from_json("full_set.json")
# Set up a server object to run the calculations using.
working_directory = "working_directory"
# Set up a backend to run the calculations on. This assume running
# on a HPC resources with the LSF queue system installed.
queue_resources = QueueWorkerResources(
number_of_threads=1,
number_of_gpus=1,
preferred_gpu_toolkit=QueueWorkerResources.GPUToolkit.CUDA,
per_thread_memory_limit=5 * unit.gigabyte,
wallclock_time_limit="05:59",
)
worker_script_commands = [
"conda activate forcebalance", "module load cuda/10.1"
]
calculation_backend = DaskLSFBackend(
minimum_number_of_workers=1,
maximum_number_of_workers=50,
resources_per_worker=queue_resources,
queue_name="gpuqueue",
setup_script_commands=worker_script_commands,
adaptive_interval="1000ms",
)
with calculation_backend:
server = EvaluatorServer(
calculation_backend=calculation_backend,
working_directory=working_directory,
port=8002,
)
with server:
# Request the estimates.
client = EvaluatorClient(ConnectionOptions(server_port=8002))
request, _ = client.request_estimate(
property_set=data_set,
force_field_source=force_field_source,
)
# Wait for the results.
results, _ = request.results(True, 5)
results.json(f"results.json")
def main():
# Set up logging
logging.basicConfig(level=logging.INFO)
output_directory = "training_sets"
os.makedirs(output_directory, exist_ok=True)
rho_pure_h_vap = PhysicalPropertyDataSet.from_json(
"../../../pure_optimisation/data_set_generation/expanded_set/training_set.json"
)
rho_pure = PhysicalPropertyDataSet.from_json(
"../../../pure_optimisation/data_set_generation/expanded_set/training_set.json"
)
rho_pure.filter_by_property_types("Density")
h_mix_rho_x = PhysicalPropertyDataSet.from_json(
"../../../mixture_optimisation/data_set_generation/"
"expanded_set/training_sets/h_mix_rho_x_training_set.json")
h_mix_rho_x_rho_pure = PhysicalPropertyDataSet()
h_mix_rho_x_rho_pure.merge(rho_pure)
h_mix_rho_x_rho_pure.merge(h_mix_rho_x)
h_mix_rho_x_rho_pure.json(
os.path.join(output_directory, "h_mix_rho_x_rho_pure.json"))
h_mix_rho_x_rho_pure.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_rho_x_rho_pure.csv"))
h_mix_rho_x_rho_pure_h_vap = PhysicalPropertyDataSet()
h_mix_rho_x_rho_pure_h_vap.merge(rho_pure_h_vap)
h_mix_rho_x_rho_pure_h_vap.merge(h_mix_rho_x)
h_mix_rho_x_rho_pure_h_vap.json(
os.path.join(output_directory, "h_mix_rho_x_rho_pure_h_vap.json"))
h_mix_rho_x_rho_pure_h_vap.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_rho_x_rho_pure_h_vap.csv"))
def main():
setup_timestamp_logging()
# Load in the force field
force_field_path = "smirnoff99Frosst-1.1.0.offxml"
force_field_source = SmirnoffForceFieldSource.from_path(force_field_path)
# Create a data set containing three solvation free energies.
data_set = PhysicalPropertyDataSet.from_json("hydration_data_set.json")
data_set.json("hydration_data_set.json", format=True)
# Set up a server object to run the calculations using.
server = setup_server(backend_type=BackendType.LocalGPU,
max_number_of_workers=1,
port=8002)
with server:
# Request the estimates.
property_estimator = EvaluatorClient(
ConnectionOptions(server_port=8002))
options = RequestOptions()
options.calculation_layers = ["SimulationLayer"]
options.add_schema("SimulationLayer", "SolvationFreeEnergy",
_get_fixed_lambda_schema())
request, _ = property_estimator.request_estimate(
property_set=data_set,
force_field_source=force_field_source,
options=options,
)
# Wait for the results.
results, _ = request.results(True, 60)
# Save the result to file.
results.json(f"results.json", True)
def main():
# Set up logging
logging.basicConfig(level=logging.INFO)
output_directory = "training_sets"
os.makedirs(output_directory, exist_ok=True)
pure_density_h_vap = PhysicalPropertyDataSet.from_json(
"../../pure_optimisation/data_set_generation/training_set.json"
)
pure_density = PhysicalPropertyDataSet.from_json(
"../../pure_optimisation/data_set_generation/training_set.json"
)
pure_density.filter_by_property_types("Density")
h_mix_v_excess = PhysicalPropertyDataSet.from_json(
"../../mixture_optimisation/data_set_generation/"
"training_sets/h_mix_v_excess_training_set.json"
)
h_mix_binary_density = PhysicalPropertyDataSet.from_json(
"../../mixture_optimisation/data_set_generation/"
"training_sets/h_mix_density_training_set.json"
)
h_mix_binary_density_pure_density = PhysicalPropertyDataSet()
h_mix_binary_density_pure_density.merge(pure_density)
h_mix_binary_density_pure_density.merge(h_mix_binary_density)
h_mix_binary_density_pure_density.json(
os.path.join(output_directory, "h_mix_binary_density_pure_density.json")
)
h_mix_binary_density_pure_density.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_binary_density_pure_density.csv")
)
h_mix_v_excess_pure_density = PhysicalPropertyDataSet()
h_mix_v_excess_pure_density.merge(pure_density)
h_mix_v_excess_pure_density.merge(h_mix_v_excess)
h_mix_v_excess_pure_density.json(
os.path.join(output_directory, "h_mix_v_excess_pure_density.json")
)
h_mix_v_excess_pure_density.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_v_excess_pure_density.csv")
)
h_mix_binary_density_pure_density_h_vap = PhysicalPropertyDataSet()
h_mix_binary_density_pure_density_h_vap.merge(pure_density_h_vap)
h_mix_binary_density_pure_density_h_vap.merge(h_mix_binary_density)
h_mix_binary_density_pure_density_h_vap.json(
os.path.join(output_directory, "h_mix_binary_density_pure_density_h_vap.json")
)
h_mix_binary_density_pure_density_h_vap.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_binary_density_pure_density_h_vap.csv")
)
h_mix_v_excess_pure_density_h_vap = PhysicalPropertyDataSet()
h_mix_v_excess_pure_density_h_vap.merge(pure_density_h_vap)
h_mix_v_excess_pure_density_h_vap.merge(h_mix_v_excess)
h_mix_v_excess_pure_density_h_vap.json(
os.path.join(output_directory, "h_mix_v_excess_pure_density_h_vap.json")
)
h_mix_v_excess_pure_density_h_vap.to_pandas().to_csv(
os.path.join(output_directory, "h_mix_v_excess_pure_density_h_vap.csv")
)
