hpmcm.utils module

hpmcm.utils.associateSourcesToFootprints(data, cluster_key, pixel_match_scale=1)

Loop through data and associate sources to clusters

Return type:

list[ndarray]

Parameters:

• data (list[DataFrame]) – Input DataFrames

• cluster_key (ndarray) – 2D-map of cluster Ids by pixel position

• pixel_match_scale (int) – Scale-factor used in making cluster map

Returns:

• Lists of clusters associated to each source

• output[i][j] will give the id of the cluster associated

• to source j in input catalog i.

hpmcm.utils.fillCountsMapFromArrays(x_locals, y_locals, n_pix, weights=None)

Fill a source counts map

Return type:

ndarray

Parameters:

• x_locals (ndarray) – Local pixel x-positions

• y_locals (ndarray) – Local pixel y-positions

• n_pix (ndarray) – Number of pixels in x,y for counts map

• weights (ndarray | None) – If provided, weights to apply for each entry in counts map

Return type:

Counts map of source in cell, projected into n_pix,n_pix grid

hpmcm.utils.fillCountsMapFromDf(df, n_pix, weight_name=None, pixel_match_scale=1)

Fill a source counts map from a reduced dataframe for one input catalog

Return type:

ndarray

Parameters:

• df (DataFrame) – DataFrame with local pixel positions (x_cell, y_cell)

• n_pix (ndarray) – Number of pixels in x,y for counts map

• weight_name (str | None) – If provided column to use for weights

• pixel_match_scale (int) – Scale-factor to use in making cluster map

Return type:

Counts map of source in cell, projected into n_pix,n_pix grid

hpmcm.utils.findClusterIds(df, cluster_key, pixel_match_scale=1)

Associate sources to clusters using clusterkey which is a map where any pixel associated to a cluster has the cluster index as its value

Return type:

ndarray

Parameters:

• df (DataFrame) – DataFrame with local pixel positions (x_cell, y_cell)

• cluster_key (ndarray) – 2D-map of cluster Ids by pixel position

• pixel_match_scale (int) – Scale-factor to use in making cluster map

Return type:

Ids of associated clusters

hpmcm.utils.findClusterIdsFromArrays(x_locals, y_locals, cluster_key)

Associate sources to clusters using clusterkey which is a map where any pixel associated to a cluster has the cluster index as its value

Return type:

ndarray

Parameters:

• x_locals (ndarray) – Local pixel x-positions

• y_locals (ndarray) – Local pixel y-positions

• cluster_key (ndarray) – 2D-map of cluster Ids by pixel position

Return type:

Ids of associated clusters

hpmcm.utils.getFootprints(counts_map, buf, pixel_match_scale=1)

Take a source counts map and do clustering using Footprint detection

Return type:

dict

Parameters:

• counts_map (ndarray) – Map of source counts

• buf (int) – Number of pixels in cell-edge buffer

• pixel_match_scale (int) – Scale-factor used in making cluster map

Returns:

• Footprint data

• +—————+——————+———————————+

• | Key | Type | Description |

• +===============+==================+=================================+

• | image | np.ndarray | Counts map of sources |

• +—————+——————+———————————+

• | footprints | FootprintSet | Clustering footprints |

• +—————+——————+———————————+

• | footprint_key | np.ndarray | Array with cluster associations |

• +—————+——————+———————————+

hpmcm.utils.reduceObjectTable(basefile, outfile, extra_cols=None)

Reduce an object table to just the colums needed for matching

Return type:

None

Parameters:

• basefile (str) – Original file name

• outfile (str) – Output file name

• extra_cols (list[str] | None) – Extra columns to copy

Notes

This will produce a DataFrame with at least these columns:

Column name	Description
id	source ID
tract	Tract source was found in
patch	Patch source was found in
ra	RA in degrees
dec	DEC in degress
snr	Signal-to-Noise of source, used for filtering and centroiding
{band}_gaapPsfFlux	Flux, for band in u,g,r,i,z,y
{band}_gaapPsfFluxErr	Flux error, for band in u,g,r,i,z,y