Handling parameters in ParaQeet

The OptimizationMap is a utility class that collects all parameters that shall be considered during optimization and associates them with the corresponding Optimizable interface. With this class, Quantity objects can be traced back to the Optimizable to which they belong. Before optimization, an instance of this class needs to be filled and passed to the optimizer.

from paraqeet.optimizable import Optimizable
from paraqeet.optimization_map import OptimizationMap
from paraqeet.quantity import Quantity
from paraqeet.signal.envelopes import ConstantEnvelope, FlatTopGaussianEnvelope
from paraqeet.signal.iq_mixer import IQMixer

Example devices and their parameters

We define a tone and a signal generator and look at their parameters:

tone = ConstantEnvelope()

gen = IQMixer(envelopes=[tone])

We note that both tone and gen are instances of Optimizable.

print(isinstance(tone, Optimizable))
print(isinstance(gen, Optimizable))

True
True

As such they must have a get_parameters method implemented. We can obtain the (default) parameters of the tone as

params_tone = tone.get_parameters()
print(params_tone)

[Amplitude: 24.7 MHz x 2pi, t_final: 32 ns]

and those of the generator as

params_gen = gen.get_parameters()
print(params_gen)

[Amplitude: 24.7 MHz x 2pi, t_final: 32 ns, lo_freq: 4.8 GHz x 2pi, Phase: 0 rad]

As we see, in this case, gen has the parameters of the tone as its parameters. Moreover, it has two additional parameters, namely, the frequency of the local oscillator (‘lo_freq’) and the phase of the pulse (‘Phase’). We also note that the get_parameters returns a list of ParaQeet’s Quantity objects:

print(isinstance(params_gen, list))
print([isinstance(param, Quantity) for param in params_gen])

True
[True, True, True, True]

The Optimization Map

To handle the parameters of both tones, we make an OptimizationMapand add the parameters of the gen explcitely.

optmap = OptimizationMap()
optmap.add(gen, params_gen)

Note that we also need to pass gen itself when we add the corresponding parameters to the optmap. We can get a list output of all parameters with

optmap.get_all_parameters()

[Amplitude: 24.7 MHz x 2pi,
 t_final: 32 ns,
 lo_freq: 4.8 GHz x 2pi,
 Phase: 0 rad]

If we just want to optimize the frequency, we set

optmap.add(gen, [params_gen[2]])
optmap.get_all_parameters()

[lo_freq: 4.8 GHz x 2pi]

which overwrites the previous parameters associated with gen.

Adding more devices

As a second drive, we create a signal shaped by a flat-top Gaussian envelope:

tone_2 = FlatTopGaussianEnvelope()
gen_2 = IQMixer(envelopes=[tone_2])

If we don’t specify an explicit list of parameters, all of them get added.

optmap.add(gen_2)
optmap.get_all_parameters()

[lo_freq: 4.8 GHz x 2pi,
 Amplitude: 24.7 MHz x 2pi,
 t_up: 6.4 ns,
 t_down: 25.6 ns,
 ramp_time: 3.2 ns,
 lo_freq: 4.8 GHz x 2pi,
 Phase: 0 rad]

If we just print optmap we can also see that the parameters are associated with different objects

print(optmap)

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[lo_freq: 4.8 GHz x 2pi]

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi, t_up: 6.4 ns, t_down: 25.6 ns, ramp_time: 3.2 ns, lo_freq: 4.8 GHz x 2pi, Phase: 0 rad]

Selecting parameters

There is a convenient filter method to select parameters based on properties. The following example selects all amplitudes:

def all_amplitudes(par):
    """Get the amplitudes."""
    return par.get_name() == "Amplitude"


optmap.filter_parameters(all_amplitudes)
optmap.get_all_parameters()

[Amplitude: 24.7 MHz x 2pi]

print(optmap)

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi]

Adding back all parameters:

optmap.add(gen)
optmap.add(gen_2)
print(optmap)

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi, t_up: 6.4 ns, t_down: 25.6 ns, ramp_time: 3.2 ns, lo_freq: 4.8 GHz x 2pi, Phase: 0 rad]

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi, t_final: 32 ns, lo_freq: 4.8 GHz x 2pi, Phase: 0 rad]

Now, we select every parameter with unit “Hz”:

def hz_filter(par):
    """Get every parameter with the unit 'Hz'."""
    return par.get_unit() == "Hz"


optmap.filter_parameters(hz_filter)
print(optmap)

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi, lo_freq: 4.8 GHz x 2pi]

==== <class 'paraqeet.signal.iq_mixer.IQMixer'> ====
[Amplitude: 24.7 MHz x 2pi, lo_freq: 4.8 GHz x 2pi]

In the following notebooks we will put to practice what we learned in this notebook and use the OptimizationMap to select the parameters to optimize in optimal control tasks.