atomsmltr.simulation.simulator package#

simulator#

Here we implement the Simulation class, that allows to run simulations on a configuration defined via the Configuration class.

Examples

Run a simulation with one initial condition vector

# ... init a config object with the `Configuration` class

# - import a simulation class
from atomsmltr.simulation import ScipyIVP_3D

# - init and setup
sim = ScipyIVP_3D(method="Radau")
sim.config = config

# - parameters
u0 = (0, 0, -0.15, 0, 0, 200)
t = np.linspace(0, 0.05, 1000)

# - integrate
res = sim.integrate(u0, t)

Run a batch of simulations

# ... init a config object with the `Configuration` class

# - import a simulation class
from atomsmltr.simulation import ScipyIVP_3D

# - init and setup
sim = ScipyIVP_3D(method="Radau")
sim.config = config

# - parameters
# initial conditions
vz_list = np.linspace(10, 300, 40)
u0_list = [(0, 0, -0.15, 0, 0, v) for v in vz_list]
sim.u0_list = u0_list
# time
t = np.linspace(0, 0.05, 1000)

# - run a batch in parallel
res_list = sim.run(t, npools=5, verbose=True)
class atomsmltr.simulation.simulator.Euler(config: Configuration = None)[source]#

Bases: CustomSimulationBase

A homemade simulator based on Euler’s method

Parameters:

config (Configuration, optional) – the configuration to consider for the simulation

References

TODO: put here

class atomsmltr.simulation.simulator.EulerSt(config: Configuration = None, enable_fluct: bool = True)[source]#

Bases: CustomSimulationBase

A homemade simulator based on simple Euler integration method, taking into account spontaneous emission.

Parameters:

  • config (Configuration, optional) – the configuration to consider for the simulation

  • enable_fluct (Boolean, optional (default=True)) – if set to True, fluctations are enabled. Otherwise, the simulator boils down to a deterministic Euler simulator

References

TODO: put here

class atomsmltr.simulation.simulator.RK4(config: Configuration = None)[source]#

Bases: CustomSimulationBase

A homemade simulator based on fourth order Runge-Kutta method

Parameters:

config (Configuration, optional) – the configuration to consider for the simulation

References

https://en.wikipedia.org/wiki/Runge%E2%80%93Kutta_methods

class atomsmltr.simulation.simulator.RK4St(config: Configuration = None)[source]#

Bases: CustomSimulationBase

A homemade simulator based on fourth order Runge-Kutta method, taking into account spontaneous emission.

Parameters:

config (Configuration, optional) – the configuration to consider for the simulation

References

https://en.wikipedia.org/wiki/Runge%E2%80%93Kutta_methods

class atomsmltr.simulation.simulator.ScipyIVP_3D(config: Configuration = None, method: str = 'Radau', **solve_ivp_args)[source]#

Bases: Simulation

A simulation class based on Scipy’s solve_ivp solver

Parameters:

  • config (Configuration, optional) – the configuration to consider for the simulation

  • method (str, optional) – method used for the solve_ivp solver, by default “Radau”

  • **solve_ivp_args – all other arguments are directly passed to solve_ivp

References

https://docs.scipy.org/doc/scipy/reference/generated/scipy.integrate.solve_ivp.html

dudt(t, u)[source]#

should return the derivative of the position/speed vector u

get_force(u)[source]#

returns the force felt at a position/speed vector u

Parameters:

u (array, shape (6,) or (n1, n2, .., 6)) – array of cartesian coordinates (position and speed) in the lab frame

Returns:

force – the force at the coordinates given by pos_speed_vector

Return type:

array, shape (3,) or (n1, n2, .., 3)

class atomsmltr.simulation.simulator.Simulation(config: Configuration = None)[source]#

Bases: ABC

The generic Simulation object

Parameters:

config (Configuration, optional) – the configuration to consider, by default None

Note

this is an abstract class, actual implementations are defined elsewhere and inherit from this class

property config#

the configuration for this simulation

Type:

Configuration

abstractmethod dudt(t, u)[source]#

should return the derivative of the position/speed vector u

abstractmethod get_force(u: ndarray) → ndarray[source]#

returns the force felt at a position/speed vector u

Parameters:

u (array, shape (6,) or (n1, n2, .., 6)) – array of cartesian coordinates (position and speed) in the lab frame

Returns:

force – the force at the coordinates given by pos_speed_vector

Return type:

array, shape (3,) or (n1, n2, .., 3)

integrate(u0: ndarray, t: ndarray)[source]#

Integrates the system with initial conditions u0

Parameters:

  • u0 (array, shape (6,)) – the initial conditions (x, y, z, vx, vy, vz)

  • t (array, shape (n,)) – the timesteps to integrate

Returns:

the result of the simulation

Return type:

res

run(t: ndarray, u0_list: list = None, npools: int = 0, verbose: bool = False) → list[source]#

Runs a batch of simulations from a list of initial conditions

Parameters:

  • t (array, shape (n,)) – time steps for the simulation

  • u0_list (list, optional) – list of initial conditions, by default None

  • npools (int, optional) – number of pools for parallel computing. If set to zero, no paralalelisation, by default 0

  • verbose (bool, optional) – if set to True, a progress bar is displayed, by default False

Returns:

res_list – a list of results

Return type:

list

Examples

# ... init a config object with the `Configuration` class

# - import a simulation class
from atomsmltr.simulation import ScipyIVP_3D

# - init and setup
sim = ScipyIVP_3D(method="Radau")
sim.config = config

# - parameters
# initial conditions
vz_list = np.linspace(10, 300, 40)
u0_list = [(0, 0, -0.15, 0, 0, v) for v in vz_list]
sim.u0_list = u0_list
# time
t = np.linspace(0, 0.05, 1000)

# - run a batch in parallel
res_list = sim.run(t, npools=5, verbose=True)
property u0_list#

a list of initial conditions for batch running

Type:

list

class atomsmltr.simulation.simulator.VelocityVerlet(config: Configuration = None)[source]#

Bases: CustomSimulationBase

A homemade simulator based on velocity verlet method

Parameters:

config (Configuration, optional) – the configuration to consider for the simulation

References

TODO: put here

Submodules#