DiaSourceComparison

class lsst.analysis.ap.DiaSourceComparison(summary: DataFrame, matched: DataFrame, ref_only: DataFrame, new_only: DataFrame, flag_changes: DataFrame)

Bases: object

The diaSources of two runs, matched by position.

Attributes

summarypandas.DataFrame

One row per instrument and visit, with the band of the visit (band), the number of sources in each run (n_ref, n_new), matched (n_matched) and in one run only (n_ref_only, n_new_only), and the dataset each run’s sources were read from (table_ref, table_new; None if the run has no sources for the visit).

matchedpandas.DataFrame

One row per matched pair: the IDs and detectors of both sources, their separation (separation_arcsec), the pixel offset of the new source (dx, dy; NaN if the detectors differ), each compared column with _ref and _new suffixes, the PSF flux difference in units of its uncertainty (delta_psfFlux_sigma), and the boolean columns whose values differ (flags_changed, comma separated; isNegative if the sign changed).

ref_only, new_onlypandas.DataFrame

Sources found in one run only, with all their columns plus reason, and other_diaSourceId, other_separation_arcsec and other_reliability for the source the reason refers to (the ID is that catalog’s, not a diaSourceId, for a removed detection).

flag_changespandas.DataFrame

One row per boolean column that differs in at least one pair, with the number of pairs where it is set in the reference run only (n_ref_only) and in the new run only (n_new_only).

Attributes Summary

flag_changes
matched
new_only
ref_only
summary

Methods Summary

reason_counts()

Count the unmatched sources of each run by reason.

Attributes Documentation

flag_changes: DataFrame = <dataclasses._MISSING_TYPE object>

matched: DataFrame = <dataclasses._MISSING_TYPE object>

new_only: DataFrame = <dataclasses._MISSING_TYPE object>

ref_only: DataFrame = <dataclasses._MISSING_TYPE object>

summary: DataFrame = <dataclasses._MISSING_TYPE object>

Methods Documentation

reason_counts()

Count the unmatched sources of each run by reason.

Returns

countspandas.DataFrame

Indexed by reason, with columns ref_only and new_only.