mindspore.ops.PMEEnergy
- class mindspore.ops.PMEEnergy(*args, **kwargs)[source]
Calculate the Coulumb energy of the system using PME method.
\[E = sum_{ij} q_iq_j/r_{ij}\]- Parameters
atom_numbers (int32) – the number of atoms, n.
excluded_numbers (int32) – the length of excluded list, E.
beta (float32) – the PME beta parameter, determined by the non-bond cutoff value and simulation precision tolerance.
fftx (int32) – the number of points for Fourier transform in dimension X.
ffty (int32) – the number of points for Fourier transform in dimension Y.
fftz (int32) – the number of points for Fourier transform in dimension Z.
box_length_0 (float32) – the value of boxlength idx 0
box_length_1 (float32) – the value of boxlength idx 1
box_length_2 (float32) – the value of boxlength idx 2
- Inputs:
uint_crd (Tensor, uint32) - [n, 3], the unsigned int coordinates value of each atom.
charge (Tensor, float32) - [n,], the charge carried by each atom.
nl_numbers - (Tensor, int32) - [n,], the each atom.
nl_serial - (Tensor, int32) - [n, 800], the neighbor list of each atom, the max number is 800.
scaler (Tensor, float32) - [3,], the scale factor between real space coordinates and its unsigned int value.
excluded_list_start (Tensor, int32) - [n,], the start excluded index in excluded list for each atom.
excluded_list (Tensor, int32) - [E,], the contiguous join of excluded list of each atom. E is the number of excluded atoms.
excluded_atom_numbers (Tensor, int32) - [n,], the number of atom excluded in excluded list for each atom.
- Outputs:
reciprocal_ene (float32) - the reciprocal term of PME energy.
self_ene (float32) - the self term of PME energy.
direct_ene (float32) - the direct term of PME energy.
correction_ene (float32) - the correction term of PME energy.
- Supported Platforms:
GPU