sponge.colvar.atoms.atoms 源代码

# 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.
# ============================================================================
"""
Atoms
"""

from typing import Union, List
from numpy import ndarray

import mindspore as ms
import mindspore.numpy as msnp
from mindspore.ops import functional as F
from mindspore.common import Tensor

from ..colvar import Colvar
from ...function import functions as func
from ...function import get_ms_array, get_integer, Units

__all__ = [
    'AtomsBase',
    'Atoms',
    'BatchedAtoms',
]


[文档]class AtomsBase(Colvar): r""" Base class for specific atoms group, used as the "atoms group module" in MindSPONGE. The `AtomsBase` Cell is a special subclass of `Colvar`. It has the shape `(a_1, a_2, ... , a_n, D)`, where `D` is the dimension of the atomic coordinates (usually 3). As with the Colvar Cell, when it takes as input coordinates of shape `(B, A, D)`, it returns the shape of the Tensor with an extra dimension `B`, i.e. `(B, a_1, a_2, ... , a_n, D)`. B means Batchsize, i.e. number of walkers in simulation. {a_i} means Dimensions of the Atoms Cell. Args: keep_in_box (bool): Whether to keep the coordinate in PBC box. Default: ``False``. dimension (int): Spatial dimension of the simulation system. Default: 3 name (str): Name of the Colvar. Default: 'atoms' Supported Platforms: ``Ascend`` ``GPU`` """ def __init__(self, keep_in_box: bool = False, dimension: int = 3, name: str = 'atoms' ): super().__init__( shape=(1, dimension), periodic=False, name=name, ) self.dimension = get_integer(dimension) self.keep_in_box = keep_in_box @property def ndim(self) -> int: """rank (number of dimensions) of the atoms""" return self._ndim @property def shape(self) -> tuple: """shape of the atoms""" return self._shape
[文档] def set_dimension(self, dimension: int = 3): """set the spatial dimension of the simulation system""" self.dimension = get_integer(dimension) self._shape = self.shape[:-1] + (self.dimension,) return self
[文档] def get_unit(self, units: Units = None) -> str: """return unit of the collective variables""" return units.length_unit_name
[文档] def reshape(self, input_shape: tuple): """rearranges the shape of atoms""" if input_shape != self._shape: self.do_reshape = True self._shape = input_shape self._ndim = len(self._shape) self._periodic = msnp.full(self._shape, False) return self
[文档] def coordinate_in_pbc(self, coordinate: Tensor, pbc_box: Tensor = None) -> Tensor: """displace the coordinate in PBC box.""" if pbc_box is None: return coordinate return func.coordinate_in_pbc(coordinate, pbc_box)
def construct(self, coordinate: Tensor, pbc_box: Tensor = None): r"""get position coordinate(s) of specific atom(s) Args: coordinate (Tensor): Tensor of shape (B, A, D). Data type is float. Position coordinate of atoms in system 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, ..., D). Data type is float. """ # (B, a_1, a_2, ..., a_n, D) raise NotImplementedError def _set_shape(self, shape: tuple): """set shape of Atoms""" super()._set_shape(shape) self.dimension = self._shape[-1] return self
[文档]class Atoms(AtomsBase): r""" Specific atoms group initialized using an array of atomic indices. It is a subclass of `AtomsBase`. When initializing, the Atoms Cell accepts as input an array of atomic indices, which can either be common to all walkers or have a separate index for each walker. To set a common atomic index, set `batched` to `False`, where the shape of the `index` is the same as the shape of the `Atoms` Cell, which is `(a_1, a_2, ... , a_n)`, while the shape of the returned Tensor is `(B, a_1, a_2, ... , a_n, D)`. `B` means Batchsize, i.e. number of walkers in simulation. `{a_i}` means Dimensions of the Atoms Cell. `D` means Dimension of the simulation system. Usually is 3. To set a separate atomic index for each walker, set `Batched` to `True`. In this case, the shape of `index` should be `(B, a_1, a_2, ... , a_n)`, while the shape of the `Atoms` Cell would be `(a_1, a_2, ... , a_n)`. The batch size `B` of the atomic indices should be the same as the batch size of the simulation system. The shape of the returned Tensor of the `Atoms` Cell is `(B, a_1, a_2, ... , a_n, D)`. Args: index (Union[Tensor, ndarray, List[int]]): Array of the indices of specific atoms. The shape of tensor is (a_1, a_2, ..., a_n) or (B, a_1, a_2, ..., a_n), and the data type is int. 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``. dimension (int): Spatial dimension of the simulation system. Default: 3 name (str): Name of the Colvar. Default: 'atoms' Supported Platforms: ``Ascend`` ``GPU`` Examples: >>> import mindspore as ms >>> import numpy as np >>> from mindspore import Tensor >>> from sponge.colvar import Atoms >>> crd = Tensor(np.random.random((4, 3)), ms.float32) >>> crd Tensor(shape=[4, 3], dtype=Float32, value= [[ 2.47492954e-01, 9.78153408e-01, 1.44034222e-01], [ 2.36211464e-01, 3.35842371e-01, 8.39536846e-01], [ 8.82235169e-01, 5.98322928e-01, 6.68052316e-01], [ 7.17712820e-01, 4.72498119e-01, 1.69098437e-01]]) >>> atom_1 = Atoms(0) >>> atom_1(crd) Tensor(shape=[1, 3], dtype=Float32, value= [[ 2.47492954e-01, 9.78153408e-01, 1.44034222e-01]]) """ def __init__(self, index: Union[Tensor, ndarray, List[int]], batched: bool = False, keep_in_box: bool = False, dimension: int = 3, name: str = 'atoms', ): super().__init__( keep_in_box=keep_in_box, dimension=dimension, name=name, ) self.index = get_ms_array(index, ms.int32) if self.index.ndim == 0: self.index = self.index.reshape((1,)) if batched: # shape of self.index: (B, a_1, a_2, ..., a_n) if self.index.ndim == 1: raise ValueError('The rank of index must be larger than 1 ' 'when using batched!') else: # (1, a_1, a_2, ..., a_n) <- (a_1, a_2, ..., a_n) self.index = F.expand_dims(self.index, 0) # (a_1, a_2, ..., a_n, -1) self._set_shape(self.index.shape[1:] + (self.dimension,)) self.keep_in_box = keep_in_box
[文档] def reshape(self, input_shape: tuple): """rearranges the shape of atoms""" shape = self._shape[0] + input_shape[:-1] self.index = F.reshape(self.index, shape) self._shape = self.index.shape[1:] + (self.dimension,) self._ndim = self.index.ndim self._periodic = msnp.full(self._shape, False) return self
def construct(self, coordinate: Tensor, pbc_box: Tensor = None): r"""get position coordinate(s) of specific atom(s) Args: coordinate (Tensor): Tensor of shape (B, A, D). Data type is float. Position coordinate of atoms in system. `A` means Number of atoms in system. 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. """ # (B, a_1, a_2, ..., a_{n}, D) <- (B, A, D) atoms = func.gather_vector(coordinate, self.index) if self.keep_in_box: atoms = self.coordinate_in_pbc(atoms, pbc_box) return atoms
[文档]class BatchedAtoms(Atoms): r""" A batched version of Atoms Cell. It is a subclass of `Atoms`. the shape of `index` should be `(B, a_1, a_2, ... , a_n)`, while the shape of the `Atoms` Cell would be `(a_1, a_2, ... , a_n)`. The batch size `B` of the atomic indices should be the same as the batch size of the simulation system. The shape of the returned Tensor of the `Atoms` Cell is `(B, a_1, a_2, ... , a_n, D)`. `{a_i}` means Dimensions of the Atoms Cell. `D` means Dimension of the simulation system. Usually is 3. Args: index (Union[Tensor, ndarray, List[int]]): Array of the indices of specific atoms. The shape of tensor is (B, a_1, a_2, ..., a_{n}), and the data type is int keep_in_box (bool): Whether to displace the coordinate in PBC box. Default: ``False``. dimension (int): Spatial dimension of the simulation system. Default: 3 name (str): Name of the Colvar. Default: 'atoms' Supported Platforms: ``Ascend`` ``GPU`` """ def __init__(self, index: Union[Tensor, ndarray, List[int]], keep_in_box: bool = False, dimension: int = 3, name: str = 'atoms', ): super().__init__( index=index, batched=True, keep_in_box=keep_in_box, dimension=dimension, name=name, )