lofar-sources

LOFAR source classification

requires pygraphviz and graphviz to plot the flowchart

function make_sample can be used to generate a random subsample of a given mask, to be used in making cutout images for inspection with code from https://github.com/mhardcastle/lgz

Columns added to the catalogue:

• Ng - the number of Gaussians making up the source
• cLR (has nan)
• LR - the likelihood ratio (nan's replaced with 0)

flags

• artefact - flag 1 for artefacts

Nearest neighbour (NN) details:

• NN_LR - likelihood ratio of NN
• NN_sep - distance (in arcsec) to NN
• NN_idx - catalogue index of NN
• NN5_sep - distance to the 4th nearest neighbour
• NN_Total_flux - flux of NN
• NN_Frat - flux ratio of source to NN
• NN_Maj - size of NN

Workflow:

• find_artefact_candidates.py -- find candidate artefacts
• visual confirmation of artefacts with classify.py from lgz
• lofar_source_sorter.py --
• get_visual_flags.py -- apply flags generated from several passes of classify.py on different catagories
• write_catalogues.py -- write final catalogue, merging outputs from LGZ, ML, large galaxy strands

misc other scripts

• fix_2mass_catalogue.py -- manually fix some very wrong sizes in the 2MASX catalogue
• get_large_galaxies.py -- produce catalogues of the sources associated with very large optical galaxies
• get_intersecting_sources.py -- flag the sources that intersect/overlap
• make_summary.py -- combine inspection plots into single pdf file