---
title: CurveFitObservables (latest version)
description: API reference for qiskit_noise_learning.analysis.CurveFitObservables in the latest version of qiskit-noise-learning
source: https://quantum.cloud.ibm.com/docs/en/api/qiskit-noise-learning/generated/analysis-curve-fit-observables
---

# qiskit\_noise\_learning.analysis.CurveFitObservables

*class* `qiskit_noise_learning.analysis.CurveFitObservables`

[GitHub](https://github.com/Qiskit/qiskit-noise-learning/tree/stable/0.1/qiskit_noise_learning/analysis/curve_fit_observables.py)

Bases: [`AnalysisStage`](/docs/api/qiskit-noise-learning/generated/analysis-analysis-stage "qiskit_noise_learning.analysis.analysis_pipeline.AnalysisStage")

Fit observable data to exponential decays of the form `a * f**fragment_depth`, and average any remaining observables over randomizations.

This stage will curve fit data for any unbound paths in `fit.paths`. If `fit.paths is None`, any path in the data with multiple fragment depths will be curve fit. In both cases, any remaining paths will be averaged.

Each curve-fit row carries the following per-row metadata, which the averaged rows do not have:

- `"spam_fidelity"`, `"spam_fidelity_std"`: the fitted prefactor $a$ and its `1`-sigma uncertainty.
- `"chi_squared"`: the raw chi-squared of the fit.
- `"reduced_chi_squared"`: the raw chi-squared divided by the degrees of freedom, that is the number of fragment depths minus the two fit parameters, or `nan` if there are none.

### \_\_init\_\_

`__init__()`

## Methods

|                                                                                                                                            |                                                                                                                                                                       |
| ------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [`__init__`](#qiskit_noise_learning.analysis.CurveFitObservables.__init__ "qiskit_noise_learning.analysis.CurveFitObservables.__init__")() |                                                                                                                                                                       |
| `run`(fit)                                                                                                                                 | Run this stage, returning a new [`Fit`](/docs/api/qiskit-noise-learning/generated/analysis-fit "qiskit_noise_learning.analysis.Fit") with the output level populated. |

## Attributes

|                                                                                                                                                      |                                   |
| ---------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------- |
| [`input_level`](#qiskit_noise_learning.analysis.CurveFitObservables.input_level "qiskit_noise_learning.analysis.CurveFitObservables.input_level")    | The data level this stage reads.  |
| [`output_level`](#qiskit_noise_learning.analysis.CurveFitObservables.output_level "qiskit_noise_learning.analysis.CurveFitObservables.output_level") | The data level this stage writes. |

### input\_level

The data level this stage reads.

### output\_level

The data level this stage writes.