analysis_models

tergite_autocalibration.lib.utils.analysis_models

Classes:

Name	Description
`ExpDecayModel`	Generate an exponential decay model that can be fit to randomized benchmarking data.
`QuadraticModel`	A quadratic model, with three Parameters: `x0`, `a`, and `c`.
`RabiModel`	Generate a cosine model that can be fit to Rabi oscillation data.
`RamseyModel`	Model for exponentially decaying sinusoidal data of the form
`ThreeClassBoundary`	Defines the classification boundaries when discriminating between
`TwoClassBoundary`	Converts the boundary encoded in the LDA discriminator.

Functions:

Name	Description
`parabolic`	Return a parabolic function.

ExpDecayModel

ExpDecayModel(*args, **kwargs)

Bases: Model

Generate an exponential decay model that can be fit to randomized benchmarking data.

QuadraticModel

QuadraticModel(independent_vars=['x'], prefix='', nan_policy='raise', **kwargs)

Bases: Model

A quadratic model, with three Parameters: x0, a, and c.

Defined as:

.. math::

1	`f(x; x0, a, c) = a * (x-x0)**2 + c`

Methods:

Name	Description
`guess`	Estimate initial model parameter values from data.

guess

guess(data, x, **kwargs)

Estimate initial model parameter values from data.

RabiModel

RabiModel(*args, **kwargs)

Bases: Model

Generate a cosine model that can be fit to Rabi oscillation data.

RamseyModel

RamseyModel(*args, **kwargs)

Bases: Model

Model for exponentially decaying sinusoidal data of the form amplitudeexp(-tau * t)sin(frequency * t + phase) + offset tau is the characteristic decay constant and omega the frequency of the measured Ramsey Oscillations. The data are collected over a sequence of Ramsey delays t, i.e. delays between two consecutive X90 gates. Used by measurements that calibrate or characterize qubit dephasing: Ramsey correction, T2, T2echo

ThreeClassBoundary

ThreeClassBoundary(lda: LinearDiscriminantAnalysis)

Defines the classification boundaries when discriminating between the |0>, |1>, |2> qubit states. Such definition requires 5 parameters: the (I,Q) coordinates of the point where the three lines meet (centroid) the angles omega_ij that define the direction of each line with respect to the I axis

Attributes

centroid_I: float the I coordinate of the point where the three classification lines meet centroid_Q: float the Q coordinate of the point where the three classification lines meet omega_01: float the angle in degrees in the range [0,360) of the boundary between |0> and |1> omega_12: float the angle in degrees in the range [0,360) of the boundary between |1> and |2> omega_20: float the angle in degrees in the range [0,360) of the boundary between |2> and |0>

Methods

boundary_line (int: class_a, class_a) -> (np.ndarray, np.ndarray): used for plotting, returns the x and y points needed to plot the line between the classes class_a and class_b. The line starts at the centroid.

Methods:

Name	Description
`omega`	Be careful: angle defined in the [0,360) range

omega

omega(index_a: int, index_b: int)

Be careful: angle defined in the [0,360) range

TwoClassBoundary

TwoClassBoundary(lda: LinearDiscriminantAnalysis)

Converts the boundary encoded in the LDA discriminator. The LDA boundary (also called threshold) has the form Ax + By + y_intercept = 0. This boundary is converted: i. To the form y = lamda * x + y_intercept, used in plotting ii. To the form (theta, threshold) used by the Quantify Scheduler for Thresholded Aqcuisitions

Attributes

lamda: float the slope coefficient of form (i) y_intercept: float the y-axis intercept of form (i) theta_rad: float the angle of the boundary, used for form (ii) threshold: float the distance from the IQ origin to the boundary line, used for form (ii)

parabolic

parabolic(x, x0=0, a=0.0, c=0.0)

Return a parabolic function.

parabolic(x, x0, a, c) = a * (x-x0)**2 + c