quantity.targets.init_dihedral_distribution_target

quantity.targets.init_dihedral_distribution_target#

init_dihedral_distribution_target(target, bonds, nbins=150, degree=True, smoothing='gaussian', gamma=1.0, approximate_average=False)[source]#

Initializes reference and compute function for a dihedral angle dist.

Parameters:

  • key – Target key

  • target (Union[Array, ndarray, bool, number, bool, int, float, complex]) – Target distribution. Array of bin centers and corresponding probabilities.

  • bonds (Union[Array, ndarray, bool, number, bool, int, float, complex]) – Indices of the sites enclosing the dihedral angle.

  • degree (bool) – Whether the reference distribution is given in degree.

  • smoothing (str) – Method to smooth the predicted distribution.

  • nbins (int) – The number of bins of the target distribution.

  • gamma (float) – Scaling factor in the loss function.

  • approximate_average (bool) – Use a linearized approximation of the ensemble average in the reweighting context.

Return type:

TargetInit

Returns:

Returns a target dict and a compute function to be passed to the trainers.

Contents