Simulation

A DMC simulation class. Uses a number of subclasses to manage its methods

Properties and Methods

load_lib: method
reload_lib: method

__init__(self, params): 

Initializes the simulation from the simulation parameters

• params: SimulationParameters

  the parameters for the simulation

configure_simulation(self, name='dmc', description='a dmc simulation', walker_set=None, time_step=0, alpha=None, potential=None, atomic_units=False, steps_per_propagation=None, mpi_manager=True, importance_sampler=None, num_wavefunctions=0, ignore_errors=False, branching_threshold=1.0, energy_error_value=1000000000.0, max_weight_threshold=None, min_potential_threshold=None, branch_on_steps=False, parallelize_diffusion=True, branch_on_cores=False, random_seed=None, pre_run_script=None, post_run_script=None): 

• name: str

  No description…

• description: str

  No description…

• walker_set: WalkerSet

  No description…

• time_step: int

  No description…

• alpha: float

  No description…

• potential: str | Potential

  No description…

• mpi_manager: MPIManagerObject

  No description…

• steps_per_propagation: int

  No description…

• importance_sampler: ImportanceSampler

  No description…

• :returns: _

  No description…

@property
config_string(self): 

checkpoint(self, test=True): 

garbage_collect(self, test=True): 

reload(self, energies_file='energies.npy', walkers_file='walkers_{n}.npz', full_weights_file='full_weights.npy', full_energies_file='full_energies.npy'): 

Reloads the core data in a Simulation object from a checkpoint file

• core_dir: Any

  No description…

• params_file: Any

  No description…

log_print(self, *arg, allow_dummy=False, **kwargs): 

run(self): 

Runs the DMC until we’ve gone through the requested number of time steps, checkpoint-ing if there’s a crash

• :returns: _

  No description…

@property
lib(self): 

apply_branching(self, energies): 

evaluate_potential_and_branch(self, nsteps): 

propagate(self, nsteps=None): 

Propagates the system forward n steps

• nsteps: Any

  number of steps to propagate for; None means automatic

• :returns: _

  No description…

update_weights(self, energies, weights): 

Iteratively updates the weights over a set of vectors of energies

• energies: np.ndarray

  No description…

• weights: np.ndarray

  No description…

• :returns: np.ndarray

  No description…

chop_weights(eliminated_walkers, weights, parents, walkers, energies): 

branch(self, energies): 

Handles branching in the system.

• :returns: _

  No description…

descendent_weight(self): 

Calls into the walker descendent weighting if the timing is right

• :returns: _

  No description…

Examples

Edit Examples or Create New Examples
Edit Template or Create New Template
Edit Docstrings