Ctrl+K

Callbacks#

Base class#

class torchgpe.utils.callbacks.Callback[source]#

Base class for callbacks.

Before a simulation starts, it is provided with the instance of the gpe.bec2D.gas.Gas (stored in the gpe.utils.callbacks.Callback.gas variable) and with a dictionary of parameters for the simulation (stored in gpe.utils.callbacks.Callback.propagation_params)

Attributes

gas

The instance of the gpe.bec2D.gas.Gas class.

propagation_params

A dictionary of parameters for the simulation.

Methods

on_propagation_begin

Function called by the gpe.bec2D.gas.Gas class before the simulation begins

on_epoch_begin

Function called by the gpe.bec2D.gas.Gas at the beginning of each epoch

on_epoch_end

Function called by the gpe.bec2D.gas.Gas at the end of each epoch

on_propagation_end

Function called by the gpe.bec2D.gas.Gas class after the simulation ends

Implemented callbacks#

Dimensionality dependent callbacks#

class torchgpe.utils.callbacks.LInfNorm(compute_every=1, print_every=1)[source]#

Callback computing the \(L_\infty\) norm of the wavefunction

The \(L_\infty\) norm is defined as:

\[L_\infty = \text{max}_{(x,y)}|\Psi_t - \Psi_{t+\Delta t}|\]
Parameters:

  • compute_every (int) – Optional. The number of epochs after which the norm is computed. Defaults to 1.

  • print_every (int) – Optional. The number of epochs after which, if computed, the norm is also printed. Defaults to 1.

Attributes

norms

A list of the computed norms
class torchgpe.utils.callbacks.L1Norm(compute_every=1, print_every=1)[source]#

Callback computing the \(L_1\) norm of the wavefunction

The \(L_1\) norm is defined as:

\[L_1 = \sum_{(x,y)}|\Psi_t - \Psi_{t+\Delta t}| \, dx \, dy\]
Parameters:

  • compute_every (int) – Optional. The number of epochs after which the norm is computed. Defaults to 1.

  • print_every (int) – Optional. The number of epochs after which, if computed, the norm is also printed. Defaults to 1.

Attributes

norms

A list of the computed norms
class torchgpe.utils.callbacks.L2Norm(compute_every=1, print_every=1)[source]#

Callback computing the \(L_2\) norm of the wavefunction

The \(L_2\) norm is defined as:

\[L_2 = \sqrt{\sum_{(x,y)}|\Psi_t - \Psi_{t+\Delta t}|^2 \, dx \, dy}\]
Parameters:

  • compute_every (int) – Optional. The number of epochs after which the norm is computed. Defaults to 1.

  • print_every (int) – Optional. The number of epochs after which, if computed, the norm is also printed. Defaults to 1.

Attributes

norms

A list of the computed norms

2D specific callbacks#

class torchgpe.bec2D.callbacks.CavityMonitor(dispersive_cavity, save_every=1)[source]#

Callback monitoring the time dependent parameters of a dispersive cavity and its field.

During the simulation, the values of cavity detuning, pump strength and cavity field are stored. Once the simulation is finished, the saved parameters are accessible via the gpe.bec2D.callbacks.CavityMonitor.alpha, gpe.bec2D.callbacks.CavityMonitor.pump and gpe.bec2D.callbacks.CavityMonitor.cavity_detuning tensors.

Parameters:

  • dispersive_cavity (DispersiveCavity) – The cavity to be monitored.

  • save_every (int) – Optional. The number of epochs after which the parameters should be saved. Defaults to 1.

Attributes

pump

A list of the pump strengths.

cavity_detuning

A list of the cavity detunings.

alpha

A list of the cavity field amplitudes.

times

A list of the times at which the parameters were saved.
class torchgpe.bec2D.callbacks.Animation(output_file, save_every=1, fps=25, cores=1, density=True, phase=True, densityk=False, potentials=False, cavities=[], time_dependent_variables=[])[source]#

Callback generating an animation of the propagation of the wavefunction.

Parameters:

  • output_file (str) – The path where to store the mp4 animation.

  • save_every (int) – Optional. The number of epochs after which a frame of the animation is saved. Defaults to 1.

  • fps (int) – Optional. The number of frames per second of the animation. Defaults to 25.

  • cores (int) – Optional. The number of cores to use for the generation of the images. Defaults to 1.

  • density (bool) – Optional. Whether to plot the real space density. Defaults to True.

  • phase (bool) – Optional. Whether to plot the phase. Defaults to True.

  • densityk (bool) – Optional. Whether to plot the momentum space density. Defaults to False.

  • potentials (bool) – Optional. Whether to plot the potential landscape. Defaults to False.

  • cavities (list) – Optional. A list of gpe.bec2D.potentials.DispersiveCavity objects to monitor. Defaults to [].

  • time_dependent_variables (list) – Optional. A list of tuples of the form (label, function) where label is a string and function is a function of time returning a float. The value of the function will be plotted as a function of time. Defaults to [].

On this page