Python and Astropy • sagui

Sagui returns a categorical map and regional photometry tables. CSV is portable; FITS preserves the spatial WCS with the map.

Export from R

write.csv(regional$flux_long, "regional_seds_long.csv", row.names = FALSE)

label_image <- seg$cluster_map
label_image[is.na(label_image)] <- 0L
FITSio::writeFITSim(label_image, file = "region_labels.fits")

Here, zero means outside support or unassigned; positive integers are categorical region identifiers. Copy the original FITS header to preserve the WCS.

Read with Python

import pandas as pd
from astropy.io import fits

sed = pd.read_csv("regional_seds_long.csv")
labels = fits.getdata("region_labels.fits")

assert (sed["region"] > 0).all()
assert labels.ndim == 2

Use the region column, rather than row order, to join SED-fitting results back to the map.

Fit downstream

Sagui does not fit physical SED models. Pass its regional fluxes and uncertainties, together with the filters and redshift, to the fitting code.