laboneq_applications.tasks.evaluation
¶
This module defines the task for updating setup parameters.
evaluate_parameter_and_fit_r2_thresholds(old_parameter_values, new_parameter_values, fit_data, quantum_elements, *, parameter, default_parameter_threshold, default_fit_r2_threshold, evaluation_parameters=None)
¶
Evaluates the parameter and fit r2 thresholds.
We evaluate the analysis workflow results for a generic experiment. We output the evaluation flags, which is a dictionary of the form:
{"q0": {"success": False, "update": False}, ...}
For each quantum element:
If the r2 value of the fit is above the r2_threshold, then the success flag is
set to True.
If the change in one of the new_parameter_values relative to the
old_parameter_values is above a threshold, then the update flag is set to
True.
If both the success and update flags are True, then the QPU will be
updated.
Warning
This evaluation task is not compatible with all experiment workflows. This task
requires an evaluation parameter of type float and fit data that contains an
r2 value at the location fit_data[q].rsquared.
Note
In this evaluation task, we check only the r2 value of the fit to determine whether a generic experiment is successful, and we check only the change in one qubit parameter to determine whether the qubit update is significant. In practice, many other indictors can be used. We encourage the user to write their own experiment evaluation task to match their criteria for success and updates.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
old_parameter_values
|
dict[str, Any]
|
The old parameter values. |
required |
new_parameter_values
|
dict[str, Any]
|
The new parameter values. |
required |
fit_data
|
dict[str, ModelResult]
|
The fit data. |
required |
quantum_elements
|
QuantumElements
|
The quantum elements to run the experiments on. |
required |
parameter
|
str
|
The name of the parameter for which to evaluate the difference. |
required |
default_parameter_threshold
|
float
|
The default parameter threshold. If the change in the parameter value is
above this threshold, then |
required |
default_fit_r2_threshold
|
float
|
The default r2 threshold. If the r2 value of the fit is above this
threshold, then |
required |
evaluation_parameters
|
dict[str, dict[str, float]] | None
|
The evaluation parameters. The expected dictionary keys are "parameter_thresholds" and "fit_r2_thresholds". The "parameter_thresholds" is the dictionary of thresholds for the parameter differences, keyed by quantum element UID. The "fit_r2_thresholds" is the dictionary of thresholds for the r2 values of the fits, keyed by quantum element UID. |
None
|
Returns:
| Type | Description |
|---|---|
dict[str, dict[str, bool]]
|
The dictionary of evaluation flags. The keys are the quantum element UIDs and
the values are dictionaries of booleans. The |