pylephare

Python wrapper of LePhare SED fitting.

http://www.cfht.hawaii.edu/~arnouts/LEPHARE/lephare.html

See LePhare Documentation

It includes Monte Carlo generated SED spectra.

Acknowledgements

If you are using pylephare for your research, please add the following sentence in you acknowledgements:

_pylephare code has been funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement n°759194 - USNAC)

Running the Code: Simple LePhare run()

The library enables you to run LePhare directly from python.

Before we start...

Input Data formating

The code expects input data in form of a dataframe with the following column format: filtername1, filtername1.err, filtername2, filtername2.err, ... filternameN, filternameN.err, CONTEXT, Z-SPEC, STRING

where filtername{} is a known filter by LePhare with the format instrument.band, for instance sdss.u or ps1.z. The file pylephare/data/data_test.csv is an example.

Config File

LePhare request that you provide a config file. We invite you to check LePhare config documentation. An example is provided here: pylephare/config/default.config which is copied to your pylephare environment, see below.

All config file parameters could be changed later on.

pylephare environment

the first time you import a pylephare modul, this will create the pylephare working environment:

• the directory $LEPHAREWORK/pylephare
• the directory $LEPHAREWORK/pylephare/config
• copy the package default config file and parameter out config file inside $LEPHAREWORK/pylephare/config

Let's go

Let's load the example dataframe (pylephare/data/data_test.csv) and configfiles ($LEPHAREWORK/pylephare/config/default.config):

import pandas
from pylephare import io
data = pandas.read_csv(io._PACKAGE_ROOT+"/data/data_test.csv", sep=" ")
configfile = None # default file stored in your $LEPHAREWORK/pylephare/config/default.config

and let's load the LePhare class:

from pylephare import lephare
lp = lephare.LePhare(data, configfile)

Then, to run the SED fit, simply do:

fileouts = lp.run()

fileouts is a dictionary containing the fullpath of the config file used for the fit, the input data, the out results and the resulting spectra.

Changing the configuration

if you want to check what is inside the config file, simply do:

lp.config.get_config("configfile")

use the lp.config.set_value(KEY, VALUE) method to affect the configfile. you can swith on and off a given key by using the lp.config.switch_{on/off}_key(KEY) methods.

Monte Carlo Fit

A montecarlo fitting procedure has been implemented to better account for mesurement errors when deriving the SED fitted parameters by lephare.

For that, let's use only one entry of the catalog used in the first example:

import pandas
data = pandas.read_csv(io._PACKAGE_ROOT+"/data/data_test.csv", sep=" ")
target = data.iloc[0]

configfile = None # use the default one

The MCLePhare class is a wrapper that draw ndraw realisation of the data given their respective errors and runs lephare (as illustrated in the first example) on each. Here let's do 50 draws:

from pylephare import montecarlo
mclp = montecarlo.MCLePhare(target, 50)

and let's load lephare with our config file:

mclp.load_lephare(configfile)

For this example, let's say we will largely increate the magnitude errorfloor of the data. We had 5 filters, will therefore need 5 values:

mclp.lephare.config.set_intrinsic_error([0.1,0.08,0.08,0.08,0.1])

and then, let's run lephare on these 50 draws:

mclp.run()

and here is how it looks like:

mclp.show()

The results are stored as:

mclp.catout

	Z_BEST	CHI_BEST	MOD_BEST	EXTLAW_BEST	EBV_BEST	DIST_MOD_BEST	NBAND_USED	Z_SEC	CHI_SEC	MOD_SEC	...	MASS_MED	MASS_SUP	SFR_BEST	SFR_INF	SFR_MED	SFR_SUP	SSFR_BEST	SSFR_INF	SSFR_MED	SSFR_SUP
IDENT																					
0	0.0319	0.460957E+01	11	0	0.000	0.357280E+02	5	-99.0000	0.100000E+10	-999	...	0.711027E+01	0.727246E+01	-0.209793E+01	-0.209343E+01	-0.179313E+01	-0.148264E+01	-0.868508E+01	-0.871817E+01	-0.818073E+01	-0.763570E+01
1	0.0319	0.352857E+01	1	1	0.160	0.357280E+02	5	-99.0000	0.100000E+10	-999	...	0.710554E+01	0.726496E+01	-0.103552E+01	-0.195251E+01	-0.163073E+01	-0.134995E+01	-0.701936E+01	-0.844735E+01	-0.784873E+01	-0.738573E+01
2	0.0319	0.374409E+01	1	1	0.180	0.357280E+02	5	-99.0000	0.100000E+10	-999	...	0.710554E+01	0.726523E+01	-0.102300E+01	-0.198167E+01	-0.166232E+01	-0.137990E+01	-0.701936E+01	-0.850053E+01	-0.792148E+01	-0.743389E+01

Get synthetic photometry

to get the synthetic photometry for a given filter, nothing's easier do:

mclp.get_synthetic_photometry('sdss.u', restframe=True, influx=False)