HWO-MLOps: Machine Learning Operations on Exoplanet Catalogues

Data Access: The NAVO program provides access to data from NASA Astronomy Archives through the use of Python-based utilities. NAVO comprises:

• Mikulski Archive at Space Telescope (MAST)
• NASA High Energy Astrophysics Science Archive Research Center (HEASARC)
• NASA/IPAC Infrared Science Archive (IRSA)
• NASA Extragalactic Database (NED)

In this demonstration, we are accessing the HEARSARC Table Access Protocal (TAP) service to go over the available data tables and retrieving some data from the "Exoplanet Orbit Database".

We are then going the list of the available database parameters and populating a table of parameters that we need.

HEASARC Catalogue Interface accessed....

Number of services: 1

Service Description: None

Number of tables: 1032

List of available data tables:

List of parameters in the Exoplanet Orbit Database:

Exoplanets found:

Entries with NaNs removed

Number of rows: 340

Number of columns: 15

Exoplanet Clustering using Machine Learning

A possible use case relevant to this dataset is to apply machine learning to perform a Clustering operation on Star and Exoplanet radii (for example).

Here, we are using the Scikit-Learn ML framework...

Clustering Plots

Star radii: Solar radii

Exoplanet radii: Jupiter radii