sponge.control.Controller

class sponge.control.Controller(system: Molecule, control_step: int = 1, **kwargs)[source]

Base class for the controller module in MindSPONGE. The sponge.control.Controller is used in sponge.control.Updater to control the values of seven variables during the simulation process, including coordinate, velocity, force, energy, kinetics, virial and pbc_box.

Parameters

system (sponge.system.Molecule) – Simulation system.
control_step (int, optional) – Step interval for controller execution. Default: 1.
kwargs (dict) – Other parameters for extension.

Inputs:

coordinate (Tensor) - Coordinate. Tensor of shape \((B, A, D)\). Data type is float. Here \(B\) is the number of walkers in simulation, \(A\) is the number of atoms and \(D\) is the spatial dimension of the simulation system, which is usually 3.
velocity (Tensor) - Velocity. Tensor of shape \((B, A, D)\). Data type is float.
force (Tensor) - Force. Tensor of shape \((B, A, D)\). Data type is float.
energy (Tensor) - Energy. Tensor of shape \((B, 1)\). Data type is float.
kinetics (Tensor) - Kinetics. Tensor of shape \((B, D)\). Data type is float.
virial (Tensor) - Virial. Tensor of shape \((B, D)\). Data type is float.
pbc_box (Tensor) - Pressure boundary condition box. Tensor of shape \((B, D)\). Data type is float.
step (int) - Simulation step. Default: 0.

Outputs:

coordinate, Tensor of shape \((B, A, D)\). Coordinate. Data type is float.
velocity, Tensor of shape \((B, A, D)\). Velocity. Data type is float.
force, Tensor of shape \((B, A, D)\). Force. Data type is float.
energy, Tensor of shape \((B, 1)\). Energy. Data type is float.
kinetics, Tensor of shape \((B, D)\). Kinetics. Data type is float.
virial, Tensor of shape \((B, D)\). Virial. Data type is float.
pbc_box, Tensor of shape \((B, D)\). Periodic boundary condition box. Data type is float.

Supported Platforms:

Ascend GPU

Examples

>>> from mindspore import Tensor
>>> from sponge import Sponge, Molecule, ForceField, UpdaterMD, WithEnergyCell
>>> from sponge.control import Controller
>>> from sponge.callback import RunInfo
>>> system = Molecule(template='water.tip3p.yaml')
>>> potential = ForceField(system, parameters='SPCE')
>>> withenergy = WithEnergyCell(system, potential)
>>> class MyController(Controller):
...     def construct(self,
...             coordinate: Tensor,
...             velocity: Tensor,
...             force: Tensor,
...             energy: Tensor,
...             kinetics: Tensor,
...             virial: Tensor = None,
...             pbc_box: Tensor = None,
...             step: int = 0,
...             **kwargs):
...         return super().construct(coordinate, velocity/100,
...                                 force, energy, kinetics,
...                                 virial, pbc_box, step,
...                                 **kwargs)
>>> velocity = Tensor([[0.1008,0.,0.],[-0.8,0.,0.],[-0.8,0.,0.]])
>>> updater = UpdaterMD(
...     system=system,
...     time_step=1e-3,
...     velocity=velocity,
...     integrator='velocity_verlet',
...     temperature=300,
...     controller=MyController(system)
... )
>>> mini = Sponge(withenergy, optimizer=updater)
>>> run_info = RunInfo(1)
>>> mini.run(5, callbacks=[run_info])

property boltzmann: float

Boltzmann constant in current unit.

Returns: float, Boltzmann constant in current unit.

get_com(coordinate: Tensor, keepdims: bool = True)[source]

Get coordinate of center of mass.

Parameters

coordinate (Tensor) – Atomic coordinates. Tensor shape is \((B, A, D)\). Data type is float.
keepdims (bool, optional) – If this is set to True, the second axis will be left in the result as dimensions with size one. Default: True.

Returns

Tensor, the coordinate of the center of mass.: The shape is \((B, A, D)\) or \((B, D)\). Data type is float.

get_com_velocity(velocity: Tensor, keepdims: bool = True)[source]

Calculate velocity of center of mass.

Parameters

velocity (Tensor) – Velocity. The shape is \((B, A, D)\). Data type is float.
keepdims (bool) – If this is set to True, the second axis will be left in the result as dimensions with size one. Default: True.

Returns

Tensor, Tensor of the velocity of the center of mass.: The shape is \((B, A, D)\) or \((B, D)\). Data type is float.

get_kinetics(velocity: Tensor)[source]

Calculate kinetics according to velocity.

Parameters: velocity (Tensor) – Tensor of atomic velocities. Tensor shape is \((B, A, D)\). Data type is float.
Returns: Tensor, kinetics. Tensor shape is \((B, D)\). Data type is float.

get_pressure(kinetics: Tensor, virial: Tensor, pbc_box: Tensor)[source]

Calculate pressure according to kinetics, viral and PBC box.

Parameters

kinetics (Tensor) – Kinetics. Tensor shape is \((B, D)\). Data type is float.
virial (Tensor) – Virial. Tensor shape is \((B, D)\). Data type is float.
pbc_box (Tensor) – PBC box. Tensor shape is \((B, D)\). Data type is float.

Returns

Tensor, pressure. Tensor shape is \((B, D)\). Data type is float.

get_temperature(kinetics: Tensor = None)[source]

Calculate temperature according to velocity.

Parameters: kinetics (Tensor) – Kinetics. Tensor shape is \((B, D)\). Data type is float. Default: None.
Returns: Tensor, temperature. Tensor shape is \((B)\). Data type is float.

get_volume(pbc_box: Tensor)[source]

Calculate volume according to PBC box.

Parameters: pbc_box (Tensor) – Tensor of PBC box. Tensor shape is \((B, D)\). Data type is float.
Returns: Tensor, Tensor of volume. Shape is \((B)\). The data type is float.

set_degrees_of_freedom(dofs: int)[source]

Set degrees of freedom (DOFs).

Parameters: dofs (int) – Degrees of freedom.

set_time_step(dt: float)[source]

Set simulation time step.

Parameters: dt (float) – Time step.

update_coordinate(coordinate: Tensor)[source]

Update the coordinate of the simulation system.

Parameters: coordinate (Tensor) – Tensor of atomic coordinates. Tensor shape is \((B, A, D)\). Data type is float.
Returns: Tensor, has the same data type and shape as original coordinate.

update_pbc_box(pbc_box: Tensor)[source]

Update the parameter of PBC box.

Parameters: pbc_box (Tensor) – Tensor of PBC box. Tensor shape is \((B, D)\). Data type is float.
Returns: Tensor, has the same data type and shape as original pbc_box.