photod.bayes#

Attributes#

cc

Functions#

`set_column_mapping`(variable_mapping)	Set the column mapping used for this module. Updates the module-global 'cc'
`makeBayesEstimates3d`(starsData, priorGrid, globalParams)	Compute the bayes statistics for all the stars in a catalog partition.
`makeBayesPosteriors3d`(starsData, mapCatalog, globalParams)	Compute the bayes posteriors for all the stars in a DataFrame.
`getColorsAndPriorIndices`(catalog, params)
`loopOverEachStar`(starData, priorGrid, globalParams, ...)	Internal method with the logic to be run for each star.
`calculateChi2`(colors, colorsErr, locusColors)	Compute chi-squared map using provided 3D model locus.
`likeAndPrior`(Ar1d, FeH1d, Mr1d, chi2map, priorGrid, ...)	Compute the likelihood map, the 3D (Mr, FeH, Ar) prior from 2D (Mr, FeH) prior
`getMargPosteriors`(priorCube, likeCube, postCube, dMr, ...)	Get posterior information
`postProcess`(Ar1d, FeH1d, Mr1d, postCube, QrGrid, ...)	Get expectation values and uncertainties marginalize and get statistics.
`getEstimatesMeta`()	Creates an empty pd.DataFrame with the meta for the results

Module Contents#

cc = None[source]#

set_column_mapping(variable_mapping: pathlib.Path)[source]#: Set the column mapping used for this module. Updates the module-global ‘cc’ ColumnMap object.

makeBayesEstimates3d(starsData: nested_pandas.NestedFrame, priorGrid: numpy.ndarray, globalParams: photod.parameters.GlobalParams, batchSize: int = 100, returnPosteriors: bool = False)[source]#

Compute the bayes statistics for all the stars in a catalog partition.

Used for fast, large-scale processing. It leverages parallelization with JAX.

makeBayesPosteriors3d(starsData: nested_pandas.NestedFrame, mapCatalog: lsdb.catalog.map_catalog.MapCatalog, globalParams: photod.parameters.GlobalParams)[source]#

Compute the bayes posteriors for all the stars in a DataFrame.

The posterior arrays are extremely large in memory and, therefore, this method should only be used with a handful set of stars.

It does not use JAX for parallelization.

getColorsAndPriorIndices(catalog, params)[source]#

loopOverEachStar(starData, priorGrid, globalParams, returnPosteriors)[source]#: Internal method with the logic to be run for each star.

calculateChi2(colors, colorsErr, locusColors)[source]#: Compute chi-squared map using provided 3D model locus.

likeAndPrior(Ar1d, FeH1d, Mr1d, chi2map, priorGrid, priorIndices)[source]#: Compute the likelihood map, the 3D (Mr, FeH, Ar) prior from 2D (Mr, FeH) prior using uniform prior for Ar, and the chi_sq_min.

getMargPosteriors(priorCube, likeCube, postCube, dMr, dFeH, dAr)[source]#: Get posterior information

postProcess(Ar1d, FeH1d, Mr1d, postCube, QrGrid, QrIndices, margpostMr, margpostFeH, margpostAr)[source]#: Get expectation values and uncertainties marginalize and get statistics.

getEstimatesMeta()[source]#: Creates an empty pd.DataFrame with the meta for the results