lsst.pipe.tasks.diff_matched_tract_catalog Namespace Reference

Classes
class	DiffMatchedTractCatalogConfig
class	DiffMatchedTractCatalogConnections
class	DiffMatchedTractCatalogTask
class	MatchedCatalogFluxesConfig
class	MatchType
class	MeasurementType
class	MeasurementTypeInfo
class	Median
class	Percentile
class	SigmaIQR
class	SigmaMAD
class	SourceType
class	SourceTypeInfo
class	Statistic

Functions
	is_sequence_set (Sequence x)
	is_percentile (str x)
	_get_stat_name (*args)
	_get_column_name (band, *args)
	compute_stats (values_ref, values_target, errors_target, row, stats, suffixes, prefix, skip_diff=False)
	_get_columns (Dict bands_columns, Dict suffixes, Dict suffixes_flux, Dict suffixes_mag, Dict stats, ComparableCatalog target, str column_dist)

Variables
dict	DiffMatchedTractCatalogBaseTemplates

Function Documentation

_get_column_name()

lsst.pipe.tasks.diff_matched_tract_catalog._get_column_name ( band, * args )

protected

Definition at line 401 of file diff_matched_tract_catalog.py.

def _get_column_name(band, *args):
    return f"{band}_{_get_stat_name(*args)}"
 
 

_get_columns()

lsst.pipe.tasks.diff_matched_tract_catalog._get_columns ( Dict bands_columns, Dict suffixes, Dict suffixes_flux, Dict suffixes_mag, Dict stats, ComparableCatalog target, str column_dist )

protected

Get column names for a table of difference statistics.

Parameters
----------
bands_columns : `Dict` [`str`,`MatchedCatalogFluxesConfig`]
    Dict keyed by band of flux column configuration.
suffixes, suffixes_flux, suffixes_mag : `Dict` [`MeasurementType`, `str`]
    Dict of suffixes for each `MeasurementType` type, for general columns (e.g.
    coordinates), fluxes and magnitudes, respectively.
stats : `Dict` [`str`, `Statistic`]
    Dict of suffixes for each `Statistic` type.
target : `ComparableCatalog`
    A target catalog with coordinate column names.
column_dist : `str`
    The name of the distance column.

Returns
-------
columns : `Dict` [`str`, `type`]
    Dictionary of column types keyed by name.
n_models : `int`
    The number of models measurements will be made for.

Notes
-----
Presently, models must be identical for each band.

Definition at line 476 of file diff_matched_tract_catalog.py.

                 stats: Dict, target: ComparableCatalog, column_dist: str):
    """Get column names for a table of difference statistics.
 
    Parameters
    ----------
    bands_columns : `Dict` [`str`,`MatchedCatalogFluxesConfig`]
        Dict keyed by band of flux column configuration.
    suffixes, suffixes_flux, suffixes_mag : `Dict` [`MeasurementType`, `str`]
        Dict of suffixes for each `MeasurementType` type, for general columns (e.g.
        coordinates), fluxes and magnitudes, respectively.
    stats : `Dict` [`str`, `Statistic`]
        Dict of suffixes for each `Statistic` type.
    target : `ComparableCatalog`
        A target catalog with coordinate column names.
    column_dist : `str`
        The name of the distance column.
 
    Returns
    -------
    columns : `Dict` [`str`, `type`]
        Dictionary of column types keyed by name.
    n_models : `int`
        The number of models measurements will be made for.
 
    Notes
    -----
    Presently, models must be identical for each band.
    """
    # Initial columns
    columns = {
        "bin": int,
        "mag_min": float,
        "mag_max": float,
    }
 
    # pre-assign all of the columns with appropriate types
    n_models = 0
 
    bands = list(bands_columns.keys())
    n_bands = len(bands)
 
    for idx, (band, config_flux) in enumerate(bands_columns.items()):
        columns_suffix = [
            ('flux', suffixes_flux),
            ('mag', suffixes_mag),
        ]
        if idx == 0:
            n_models = len(config_flux.columns_target_flux)
        if (idx > 0) or (n_bands > 2):
            columns_suffix.append((f'color_{bands[idx - 1]}_m_{band}', suffixes))
        n_models_flux = len(config_flux.columns_target_flux)
        n_models_err = len(config_flux.columns_target_flux_err)
 
        # TODO: Do equivalent validation earlier, in the config
        if (n_models_flux != n_models) or (n_models_err != n_models):
            raise RuntimeError(f'{config_flux} len(columns_target_flux)={n_models_flux} and'
                               f' len(columns_target_flux_err)={n_models_err} must equal {n_models}')
 
        for sourcetype in SourceType:
            label = sourcetype.value.label
            # Totals would be redundant
            if sourcetype != SourceType.ALL:
                for item in (f'n_{itype}_{mtype.value}' for itype in ('ref', 'target')
                             for mtype in MatchType):
                    columns[_get_column_name(band, label, item)] = int
 
            for item in (target.column_coord1, target.column_coord2, column_dist):
                for suffix in suffixes.values():
                    for stat in stats.keys():
                        columns[_get_column_name(band, label, item, suffix, stat)] = float
 
            for item in config_flux.columns_target_flux:
                for prefix_item, suffixes_col in columns_suffix:
                    for suffix in suffixes_col.values():
                        for stat in stats.keys():
                            columns[_get_column_name(band, label, prefix_item, item, suffix, stat)] = float
 
    return columns, n_models
 
 

_get_stat_name()

lsst.pipe.tasks.diff_matched_tract_catalog._get_stat_name ( * args )

protected

Definition at line 397 of file diff_matched_tract_catalog.py.

def _get_stat_name(*args):
    return '_'.join(args)
 
 

compute_stats()

lsst.pipe.tasks.diff_matched_tract_catalog.compute_stats	(		values_ref,
			values_target,
			errors_target,
			row,
			stats,
			suffixes,
			prefix,
			skip_diff = False )

Compute statistics on differences and store results in a row.

Parameters
----------
values_ref : `numpy.ndarray`, (N,)
    Reference values.
values_target : `numpy.ndarray`, (N,)
    Measured values.
errors_target : `numpy.ndarray`, (N,)
    Errors (standard deviations) on `values_target`.
row : `numpy.ndarray`, (1, C)
    A numpy array with pre-assigned column names.
stats : `Dict` [`str`, `Statistic`]
    A dict of `Statistic` values to measure, keyed by their column suffix.
suffixes : `Dict` [`MeasurementType`, `str`]
    A dict of measurement type column suffixes, keyed by the measurement type.
prefix : `str`
    A prefix for all column names (e.g. band).
skip_diff : `bool`
    Whether to skip computing statistics on differences. Note that
    differences will still be computed for chi statistics.

Returns
-------
row_with_stats : `numpy.ndarray`, (1, C)
    The original `row` with statistic values assigned.

Definition at line 405 of file diff_matched_tract_catalog.py.

def compute_stats(values_ref, values_target, errors_target, row, stats, suffixes, prefix, skip_diff=False):
    """Compute statistics on differences and store results in a row.
 
    Parameters
    ----------
    values_ref : `numpy.ndarray`, (N,)
        Reference values.
    values_target : `numpy.ndarray`, (N,)
        Measured values.
    errors_target : `numpy.ndarray`, (N,)
        Errors (standard deviations) on `values_target`.
    row : `numpy.ndarray`, (1, C)
        A numpy array with pre-assigned column names.
    stats : `Dict` [`str`, `Statistic`]
        A dict of `Statistic` values to measure, keyed by their column suffix.
    suffixes : `Dict` [`MeasurementType`, `str`]
        A dict of measurement type column suffixes, keyed by the measurement type.
    prefix : `str`
        A prefix for all column names (e.g. band).
    skip_diff : `bool`
        Whether to skip computing statistics on differences. Note that
        differences will still be computed for chi statistics.
 
    Returns
    -------
    row_with_stats : `numpy.ndarray`, (1, C)
        The original `row` with statistic values assigned.
    """
    n_ref = len(values_ref)
    if n_ref > 0:
        n_target = len(values_target)
        n_target_err = len(errors_target) if errors_target is not None else n_ref
        if (n_target != n_ref) or (n_target_err != n_ref):
            raise ValueError(f'lengths of values_ref={n_ref}, values_target={n_target}'
                             f', error_target={n_target_err} must match')
 
        do_chi = errors_target is not None
        diff = values_target - values_ref
        chi = diff/errors_target if do_chi else diff
        # Could make this configurable, but non-finite values/errors are not really usable
        valid = np.isfinite(chi)
        values_type = {} if skip_diff else {MeasurementType.DIFF: diff[valid]}
        if do_chi:
            values_type[MeasurementType.CHI] = chi[valid]
 
        for suffix_type, suffix in suffixes.items():
            values = values_type.get(suffix_type)
            if values is not None and len(values) > 0:
                for stat_name, stat in stats.items():
                    row[_get_stat_name(prefix, suffix, stat_name)] = stat.value(values)
    return row
 
 
@dataclass(frozen=True)

is_percentile()

lsst.pipe.tasks.diff_matched_tract_catalog.is_percentile

(

str

x

)

Definition at line 53 of file diff_matched_tract_catalog.py.

def is_percentile(x: str):
    return 0 <= Decimal(x) <= 100
 
 

is_sequence_set()

lsst.pipe.tasks.diff_matched_tract_catalog.is_sequence_set

(

Sequence

x

)

Definition at line 49 of file diff_matched_tract_catalog.py.

def is_sequence_set(x: Sequence):
    return len(x) == len(set(x))
 
 

Variable Documentation

DiffMatchedTractCatalogBaseTemplates

dict lsst.pipe.tasks.diff_matched_tract_catalog.DiffMatchedTractCatalogBaseTemplates

Initial value:

=  {
    "name_input_cat_ref": "truth_summary",
    "name_input_cat_target": "objectTable_tract",
    "name_skymap": BaseSkyMap.SKYMAP_DATASET_TYPE_NAME,
}

Definition at line 57 of file diff_matched_tract_catalog.py.