Catalog and random data

The primary input for yet_another_wizz are tabular datasets which are managed by a special Catalog class. Catalogs can be created from in-memory datasets, tabular data files, and random generators.

The Catalog class

A catalog is a collection of patches of catalog data (coordinates, weights, redshifts, etc.), which are stored in a cache directory on disk:

Catalog(cache_directory, *[, max_workers])

A container for catalog data.

Upon creating a new catalog, the input data is split into patches which are stored separately in the catalog’s cache directory. A single patch and its metadata is managed by the following classes:

catalog.patch.Patch(cache_path[, center])	A single spatial patch of catalog data.
catalog.patch.Metadata(*, num_records, ...)	Container for patch meta data.

Random generators

Catalogs can also be generated randomly, e.g. to create random samples for the correlation measurements. The corresponding from_random() method accepts one of the following generators for uniform random data points:

randoms.BoxRandoms(ra_min, ra_max, dec_min, ...)	Generates random points within a right ascension / declination window.
randoms.HealPixRandoms(pix_values, *[, ...])	Generates random points within a HealPix mask.