I need to frequently convert pandas df to .fits files and also the reverse - I am given .fits files for which is easier for me to process them as pandas dfs. What is the most straightforward way for doing the conversions between these file types?
# 1.) how to convert a pandas df
# (or almost any ascii data file) to astropy table
# and save the table as .fits file
from astropy.table import Table
import pandas as pd
df = pd.read_csv('some_file.dat', delim_whitespace=True, skiprows=7, names=['A', 'B', 'C'])
t = Table.from_pandas(df)
t.write('some_file.fits')
# 2.) how to convert a fits file (or astropy table) to pandas df
dat = Table.read('some_file.fits')
df = dat.to_pandas()
# 3.) it is like 1.) but without a df as intermediate step -
# how to read a csv/ascii file directly as astropy table
# and save it as .fits
table = Table.read('some_file.dat', format='ascii.tab')
table.write('some_file.fits', overwrite=True)