Designed to enable querying and crossmatching of O(1B) sources. It addresses large-scale data processing challenges, in particular those brought up by LSST.
Built on top of Dask to efficiently scale and parallelize operations across multiple workers, it leverages the HATS data format for surveys in a partitioned HEALPix (Hierarchical Equal Area isoLatitude Pixelization) structure.
A HATS partitioning schema for Gaia DR3
Get Started
Install the latest release version of LSDB via conda.
$ conda install -c conda-forge lsdb
Or, if preferred, via pip.
$ pip install lsdb
Import the package, read two catalogs and perform their crossmatch.
>> import lsdb # Read the Gaia DR3 object catalog >> gaia = lsdb.read_hats(gaia_path) # Read the ZTF DR14 object catalog >> ztf = lsdb.read_hats(ztf_path) # Crossmatch the two catalogs >> ztf.crossmatch(gaia, n_neighbors=1, radius_arcsec=1)For advanced use cases, please have a look at the tutorials we put together.
Available HATS surveys
The Institute for Data Intensive Research in Astrophysics & Cosmology hosts a collection of survey catalogs in HATS format. Among them are Zwicky Transient Facility's (Data Release 14 and Zubercal) and Gaia (Data Release 3). They are available to the community for public use.
