PhotonTransferCurveSolveTask

class lsst.cp.pipe.PhotonTransferCurveSolveTask(*, config: PipelineTaskConfig | None = None, log: logging.Logger | LsstLogAdapter | None = None, initInputs: dict[str, Any] | None = None, **kwargs: Any)

Bases: PipelineTask

Task to fit the PTC from flat covariances.

The first task of the PTC measurement pipeline, PhotonTransferCurveMeasureTask (and assumed to have been run before this task), produced a list of PhotonTransferCurveDataset objects. Each dataset contains the mean signal and covariances of the difference image of the flat-field images taken at the same exposure time. The list also contains dummy datasets (with no measurements), whose purpose is to have the input and output dimensions of PhotonTransferCurveMeasureTask match.

This task, PhotonTransferCurveSolveTask, assembles the list of individual PTC datasets produced by PhotonTransferCurveMeasureTask into one single final PTC dataset, discarding the dummy datset as appropiate. The task fits the measured (co)variances to one of three models: any of the models described in equations 16 and 20 of Astier+19 and equation 20 with specifically fixed to 0. These options are referred to as EXPAPPROXIMATION, FULLCOVARIANCE, and FULLCOVARIANCE_NO_B in the configuration options of the task, respectively). Parameters of interest such as the gain and noise are derived from the fits. The FULLCOVARIANCE model is fitted to the full covariance data (as oppossed to the other two models, which are fit to the variance vs mean measurements only).

Astier+19: “The Shape of the Photon Transfer Curve of CCD sensors”, arXiv:1905.08677

Methods Summary

evalCovModel(mu, aMatrix, cMatrix, ...[, ...])	Computes full covariances model (Eq.
fillBadAmp(dataset, ampName)	Fill the dataset with NaNs if there are not enough good points.
fitDataFullCovariance(dataset)	Fit measured flat covariances to the full model in Astier+19 (Eq.
fitPtc(dataset[, computePtcTurnoff])	Fit the photon transfer curve to the Astier+19 approximation (Eq.
fitPtcRolloff(dataset)	Fit the photon transfer curve to the Astier+19 approximation (Eq.
funcFullCovarianceModel(params, x)	Model to fit covariances from flat fields; Equation 20 of Astier+19.
funcFullCovarianceModelNoB(params, x)	Model to fit covariances from flat fields; Equation 20 of Astier+19, with b=0 (equivalent to c=a*b=0 in this code).
initialFitFullCovariance(mu, cov, sqrtW)	Performs a crude parabolic fit of the data in order to start the full fit close to the solution, setting b=0 (c=0) in Eq.
run(inputCovariances[, camera, detId])	Fit measured covariances to different models.
runQuantum(butlerQC, inputRefs, outputRefs)	Ensure that the input and output dimensions are passed along.
subtractDistantOffset(muAtAmpMasked, ...[, ...])	Subtract distant offset from the covariance matrices.

Methods Documentation

evalCovModel(mu, aMatrix, cMatrix, noiseMatrix, gain, setBtoZero=False)

Computes full covariances model (Eq. 20 of Astier+19).

Parameters

munumpy.array, (N,)

List of mean signals (units: adu)

aMatrixnumpy.array, (M, M)

“a” parameter per flux in Eq. 20 of Astier+19 (units: 1/electron)

cMatrixnumpy.array, (M, M)

“c”=”ab” parameter per flux in Eq. 20 of Astier+19 (units: 1/electron^2)

noiseMatrixnumpy.array, (M, M)

“noise” parameter per flux in Eq. 20 of Astier+19 (units: electron^2)

gainfloat

Amplifier gain (e/adu)

setBtoZero=Falsebool, optional

Set “b” parameter in full model (see Astier+19) to zero.

Returns

covModelnumpy.array, (N, M, M)

Covariances model.

Notes

By default, computes the covModel for the mu’s stored(self.mu). Returns cov[Nmu, M, M]. The variance for the PTC is cov[:, 0, 0]. mu and cov are in adus and adus squared. To use electrons for both, the gain should be set to 1. This routine implements the model in Astier+19 (1905.08677). The parameters of the full model for C_ij(mu) (“C_ij” and “mu” in adu^2 and adu, respectively) in Astier+19 (Eq. 20) are:

• “a” coefficients (M by M matrix), units: 1/electron

• “b” coefficients (M by M matrix), units: 1/electron

• noise matrix (M by M matrix), units: electron^2

• gain, units: electron/adu

“b” appears in Eq. 20 only through the “ab” combination, which is defined in this code as “c=ab”.

fillBadAmp(dataset, ampName)

Fill the dataset with NaNs if there are not enough good points.

Parameters

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

The dataset containing the means, variances and exposure times.

ampNamestr

Amplifier name.

fitDataFullCovariance(dataset)

Fit measured flat covariances to the full model in Astier+19 (Eq. 20).

Parameters

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

The dataset containing information such as the means, (co)variances, and exposure times.

Returns

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

This is the same dataset as the input parameter, however, it has been modified to include information such as the fit vectors and the fit parameters. See the class PhotonTransferCurveDatase.

Notes

The parameters of the full model for C_ij(mu) (“C_ij” and “mu” in adu^2 and adu, respectively) in Astier+19 (Eq. 20) are:

• “a” coefficients (r by r matrix), units: 1/electron

• “b” coefficients (r by r matrix), units: 1/electron

• noise matrix (r by r matrix), units: electron^2

• gain, units: electron/adu

“b” appears in Eq. 20 only through the “ab” combination, which is defined in this code as “c=ab”.

Total number of parameters: #entries(a) + #entries(c) + #entries(noise) + 1. This is equivalent to r^2 + r^2 + r^2 + 1, where “r” is the maximum lag considered for the covariances calculation, and the extra “1” is the gain. If “b” is 0, then “c” is 0, and len(pInit) will have r^2 fewer entries.

fitPtc(dataset, computePtcTurnoff=True)

Fit the photon transfer curve to the Astier+19 approximation (Eq. 16).

Fit the photon transfer curve using the exponential approximation in Astier+19.

Sigma clipping is performed iteratively for the fit, as well as an initial clipping of data points that are more than config.initialNonLinearityExclusionThreshold away from lying on a straight line. This other step is necessary because the photon transfer curve turns over catastrophically at very high flux (because saturation drops the variance to ~0) and these far outliers cause the initial fit to fail, meaning the sigma cannot be calculated to perform the sigma-clipping.

If doModelPtcRolloff is True, a roll-off model will be added to the initial fit of the PTC to try and capture saturation effects. This will only be applied if ptcFitType=EXPAPPROXIMATION.

Parameters

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

The dataset containing the means, variances and exposure times.

computePtcTurnoffbool

Compute and save the PTC turnoff and PTC turnoff sampling error. If False, this will preserve the attributes in the input dataset.

Returns

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

This is the same dataset as the input parameter, however, it has been modified to include information such as the fit vectors and the fit parameters. See the class PhotonTransferCurveDatase.

Raises

RuntimeError

Raised if dataset.ptcFitType is None or empty.

fitPtcRolloff(dataset)

Fit the photon transfer curve to the Astier+19 approximation (Eq. 16) with a roll-off model to try and capture saturation effects.

Parameters

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

The dataset containing the means, variances and exposure times.

Returns

datasetlsst.ip.isr.ptcDataset.PhotonTransferCurveDataset

This is the same dataset as the input parameter, however, it has been modified to include information such as the fit vectors and the fit parameters. See the class PhotonTransferCurveDataset.

funcFullCovarianceModel(params, x)

Model to fit covariances from flat fields; Equation 20 of Astier+19.

Parameters

paramslist

Parameters of the model: aMatrix, CMatrix, noiseMatrix, gain (e/adu).

xnumpy.array, (N,)

Signal mu (adu)

Returns

ynumpy.array, (N,)

Covariance matrix.

funcFullCovarianceModelNoB(params, x)

Model to fit covariances from flat fields; Equation 20 of Astier+19, with b=0 (equivalent to c=a*b=0 in this code).

Parameters

paramslist

Parameters of the model: aMatrix, noiseMatrix, gain (e/adu).

xnumpy.array, (N,)

Signal mu (adu)

Returns

ynumpy.array, (N,)

Covariance matrix.

initialFitFullCovariance(mu, cov, sqrtW)

Performs a crude parabolic fit of the data in order to start the full fit close to the solution, setting b=0 (c=0) in Eq. 20 of Astier+19.

Parameters

munumpy.array, (N,)

Signal mu (adu)

covnumpy.array, (N, M, M)

Covariance arrays of size (M, M) (with M = config.maximumRangeCovariancesAstier), indexed by mean signal mu.

sqrtWnumpy.array, (N,)

Covariance weights, defined as 1./sqrt(Variances)

Returns

anumpy.array, (M, M)

“a” parameter per flux in Eq. 20 of Astier+19 (units: 1/electron).

cnumpy.array, (M, M)

“c”=”ab” parameter per flux in Eq. 20 of Astier+19. (units: 1/electron^2).

noiseMatrixnumpy.array, (M, M)

“noise” parameter per flux in Eq. 20 of Astier+19. (units: electron^2)

gainfloat

Amplifier gain (electron/adu)

run(inputCovariances, camera=None, detId=0)

Fit measured covariances to different models.

Parameters

inputCovarianceslist [lsst.ip.isr.PhotonTransferCurveDataset]

List of lsst.ip.isr.PhotonTransferCurveDataset datasets.

cameralsst.afw.cameraGeom.Camera, optional

Input camera.

detIdint

Detector ID to locate the detector in the camera and populate the lsst.ip.isr.PhotonTransferCurveDataset metadata.

Returns

resultslsst.pipe.base.Struct

The resultins structure contains:

outputPtcDatset

Final PTC dataset, containing information such as the means, variances, and exposure times (lsst.ip.isr.PhotonTransferCurveDataset).

runQuantum(butlerQC, inputRefs, outputRefs)

Ensure that the input and output dimensions are passed along.

Parameters

butlerQCQuantumContext

Butler to operate on.

inputRefsInputQuantizedConnection

Input data refs to load.

ouptutRefsOutputQuantizedConnection

Output data refs to persist.

subtractDistantOffset(muAtAmpMasked, covAtAmpMasked, covSqrtWeightsAtAmpMasked, start, degree=1)

Subtract distant offset from the covariance matrices.

Parameters

muAtAmpMaskednumpy.array

Masked mean flux array for a particular amplifier.

covAtAmpMaskednumpy.array

Masked measured covariances for a particular amplifier.

covSqrtWeightsAtAmpMaskednumpy.array

Masked inverse covariance weights for a particular amplifier.

startint, optional

The starting index to eliminate the core for the fit.

degreeint, optional

Degree of the polynomial fit.

Returns

covAtAmpMaskednumpy.array

Subtracted measured covariances for a particular amplifier.

covSqrtWeightsAtAmpMaskednumpy.array

Masked inverse covariance weights for a particular amplifier.

Notes

Ported from https://gitlab.in2p3.fr/astier/bfptc by P. Astier.

This function subtracts a distant offset from the covariance matrices using polynomial fitting. The core of the matrices is eliminated for the fit.

The function modifies the internal state of the object, updating the covariance matrices and related attributes.