Parameters
Surrogate Hessian accelerated parallel line-search: parameters
- class stalk.params.BondAngle(arg: float | tuple[ndarray, ndarray, ndarray] | list[tuple[ndarray, ndarray, ndarray]], label='a', unit='ang', error=0.0, limits=(0.0, 180.0), tol=1e-06, axes=None)
Bases:
Parameter
- class stalk.params.BondLength(arg: float | tuple[ndarray, ndarray] | list[tuple[ndarray, ndarray]], label='d', unit='A', error=0.0, limits=(0.0, inf), tol=1e-06, axes=None)
Bases:
Parameter
- class stalk.params.EffectiveVariance(samples=None, error=None, var_eff=None)
Bases:
object- add_errorbar_data(samples, error)
- add_var_eff(var_eff)
- property error_data
- get_errorbar(samples, nodes=1)
- get_samples(error, nodes=1)
- property var_eff
- class stalk.params.EffectiveVarianceMap(params: ParameterSet, var_eff: EffectiveVariance = None)
Bases:
object- add_var_eff(params: ParameterSet, var_eff: EffectiveVariance, thr=1e-06)
- get_samples(params: ParameterSet, error, nodes=1)
- property params
- scaling_map: list[ParameterSet, EffectiveVariance] = None
- class stalk.params.GeometryResult(pos, axes=None, elem=None)
Bases:
object- axes = None
- elem = None
- get_axes()
- get_elem()
- get_pos()
- get_result()
- pos = None
- rescale(scale)
- class stalk.params.LineSearchPoint(offset, value=None, error=0.0)
Bases:
object- property enabled
- property error
- property is_eqm
- property offset
- plot(ax: Axes, color='tab:blue', marker='o', linestyle='none', **kwargs)
- reset_value()
- tol = 1e-09
- property valid
The value is valid, when it is enabled and has a value
- property value
- class stalk.params.Parameter(value: float | list[float], error=0.0, label='p', unit='', limits=(-inf, inf), tol=1e-06)
Bases:
object- property error
- label: str = ''
- property limits
- shift(shift: float)
- unit: str = ''
- property value
- class stalk.params.ParameterHessian(hessian=None, structure=None, disable_limit=0.0)
Bases:
object- property U: ndarray
- compute_fdiff(pes: PesFunction, structure=None, dp=0.01, dpos_mode=False, **kwargs)
- property directions: ndarray
- property disable_limit
- property enabled
- property enabled_directions: ndarray
- property enabled_lambdas: ndarray
- property hessian
- init_hessian_array(hessian)
- property lambdas: ndarray
- property require_consistent: bool
- reset()
- property structure
- class stalk.params.ParameterMapping(forward_func=None, forward_args={}, backward_func=None, backward_args={}, dim=3)
Bases:
object- property backward
- check_params_consistency(params, tol=1e-06)
- check_pos_consistency(pos: ndarray, axes=None, tol=1e-06)
- dim = None
- property forward
- property incomplete
- map_backward(params: ndarray, **kwargs)
- map_forward(pos: ndarray, axes=None, **kwargs)
- set_backward(backward_func, backward_args={})
- set_forward(forward_func, forward_args={})
- class stalk.params.ParameterSet(params=None, params_err=None, value=None, error=0.0, label=None)
Bases:
LineSearchPoint- check_consistency()
- copy(params=None, params_err=None, label=None, offset=None)
- distance(other)
- distance2(other)
- file_path = None
- get_params_distribution(N=100)
- label = ''
- property params
- property params_err
- property params_list
- property samples
- shift_params(shifts)
- class stalk.params.ParameterStructure(pos=None, axes=None, elem=None, params=None, params_err=None, mapping=None, forward=None, forward_args={}, backward=None, backward_args={}, dim=3, value=None, error=0.0, label='', units='B', tol=1e-07, require_consistent=True)
Bases:
ParameterSet- property axes
- property backward
- property consistent
- copy(params=None, params_err=None, label=None, pos=None, axes=None, offset=None, **kwargs)
- property dim
- property elem
- property forward
- jacobian(dp=0.001)
- map_backward(params, **kwargs)
- map_forward(pos, axes=None, **kwargs)
- property mapping
- property params
- property periodic
- property pos
- pos_difference(pos_ref)
- remap_forward(forward, N=None, fraction=0.159, **kwargs)
- require_consistent = None
- set_position(pos, axes=None, translate=True)
- shift_params(shifts, dpos_mode=False)
- tol = None
- units = None
- class stalk.params.PesFunction(func, args: dict = {}, **kwargs)
Bases:
FunctionCaller- evaluate(structure: ParameterSet, sigma=0.0, add_sigma=False, var_eff_map=None, interactive=False, **kwargs)
- evaluate_all(structures: list[ParameterSet], sigmas=None, add_sigma=False, interactive=False, **kwargs)
- relax(structure: ParameterSet, **kwargs)
- class stalk.params.PesResult(value, error=0.0)
Bases:
object- add_sigma(sigma)
Add artificial white noise to the result for error resampling purposes.
- property error
- rescale(scale)
- property value
- class stalk.params.PhaseAngle(value, label='t', unit='rad', error=0.0)
Bases:
Parameter- property value
- stalk.params.angle(v0, v1, units='ang')
- stalk.params.bond_angle(r0, rc, r1, units='ang')
Return dihedral angle between 3 bodies
- stalk.params.distance(r0: ndarray, r1: ndarray) float
Return Euclidean distance between two positions
- stalk.params.interpolate_params(structure_a: ParameterSet, structure_b: ParameterSet, num_int)
- stalk.params.mean_bond_angles(triplets, tol=1e-06, axes=None, units='ang')
Return average bond angle over (presumably) identical position triplets
- stalk.params.mean_distances(pairs, tol=1e-06, axes=None)
Return average distance over (presumably) identical position pairs
- stalk.params.mean_param(params, tol=1e-06)
- stalk.params.periodic_bond_angle(r0, rc, r1, axes, units='ang')
Return dihedral angle between 3 bodies
- stalk.params.periodic_distance(r0, r1, axes)
Return minimum distance between two positions considering periodicity
- stalk.params.rotate_2d(arr_2d, ang, units='ang')