# Copyright 2021-2023 @ Shenzhen Bay Laboratory &
# Peking University &
# Huawei Technologies Co., Ltd
#
# This code is a part of MindSPONGE:
# MindSpore Simulation Package tOwards Next Generation molecular modelling.
#
# MindSPONGE is open-source software based on the AI-framework:
# MindSpore (https://www.mindspore.cn/)
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
"""
Atom group
"""
from typing import Union, List, Tuple
from numpy import ndarray
from mindspore import ops
from mindspore.ops import functional as F
from mindspore import Tensor, Parameter
from mindspore.nn import CellList
from .atoms import AtomsBase, Atoms
from ...function import get_integer
[docs]class Group(AtomsBase):
r"""
Group of atoms.
Args:
atoms (Union[List[AtomsBase], Tuple[AtomsBase]]):
List of AtomsBase. Member should be the subclass of AtomsBase.
batched (bool): Whether the first dimension of index is the batch size.
Default: ``False``.
keep_in_box (bool): Whether to displace the coordinate in PBC box.
Default: ``False``.
axis (int): Axis to concatenate the coordinates of atoms.
name (str): Name of the Colvar. Default: 'atoms_group'.
Supported Platforms:
``Ascend`` ``GPU``
"""
def __init__(self,
atoms: Union[List[AtomsBase], Tuple[AtomsBase]],
batched: bool = False,
keep_in_box: bool = False,
axis: int = 1,
name: str = 'atoms_group',
):
super().__init__(
keep_in_box=keep_in_box,
name=name,
)
if isinstance(atoms, AtomsBase):
atoms = [atoms]
elif isinstance(atoms, (Tensor, Parameter, ndarray)):
atoms = [Atoms(atoms, batched, keep_in_box)]
elif isinstance(atoms, (list, tuple)):
if set(map(type, atoms)) == {int}:
atoms = [Atoms(atoms, batched, keep_in_box)]
else:
raise TypeError(f'The type of "atoms" must be list, tuple or AtomsBase, '
f'but got: {type(atoms)}')
self.num_groups = len(atoms)
axis = get_integer(axis)
if axis in (0, -1):
raise ValueError(f'The axis ({axis}) cannot be 0 or -1!')
atoms_ = []
dim = 0
shape = None
periodic = ()
for i, a in enumerate(atoms):
if isinstance(a, (Tensor, Parameter, ndarray)):
a = Atoms(a, batched, keep_in_box)
elif isinstance(atoms, (list, tuple)):
if set(map(type, atoms)) == {int}:
atoms = [Atoms(atoms, batched, keep_in_box)]
elif not isinstance(a, AtomsBase):
raise TypeError(f'The type of elements in "atoms" must be AtomsBase, Tensor, Parameter or ndarray '
f'but got: {type(a)}')
shape_ = (1,) + a.shape
dim += shape_[axis]
shape_ = shape_[:axis] + shape_[axis+1:]
periodic += (F.expand_dims(a.periodic, 0),)
if i > 0 and shape_ != shape:
raise ValueError(f'The shape of the No.{i} AtomsBase {a.shape} cannot be '
f'concatenate with the shape of previous one: {shape}')
shape = shape_
atoms_.append(a)
self.atoms: List[AtomsBase] = CellList(atoms_)
shape = shape[:axis] + (dim,) + shape[axis:]
self._shape = shape[1:]
self._ndim = len(self._shape)
self.concat = ops.Concat(axis)
self._periodic = F.squeeze(self.concat(periodic), 0)
def construct(self, coordinate: Tensor, pbc_box: Tensor = None):
r"""get position coordinates of atoms group
Args:
coordinate (Tensor): Tensor of shape (B, A, D). Data type is float.
Position coordinate of atoms in system.
`B` means batchsize, i.e. number of walkers in simulation.
`A` means number of colvar in system.
`D` means dimension of the simulation system. Usually is 3.
pbc_box (Tensor): Tensor of shape (B, D). Data type is float.
Tensor of PBC box. Default: ``None``.
Returns:
position (Tensor): Tensor of shape (B, a_1, a_2, ..., a_n, D). Data type is float.
`a_{i}` means Dimension of specific atoms.
"""
atoms = ()
for i in range(self.num_groups):
# (B, a_1'(i), a_2, ..., a_n, D)
atoms += (self.atoms[i](coordinate, pbc_box),)
# (B, a_1, a_2, ..., a_n, D) <- (B, a_1'(i), a_2, ..., a_n, D)
atoms = self.concat(atoms)
if pbc_box is not None and self.keep_in_box:
atoms = self.coordinate_in_pbc(atoms, pbc_box)
if self.do_reshape:
new_shape = coordinate.shape[0] + self._shape
atoms = F.reshape(atoms, new_shape)
return atoms