Installing frb

This document describes how to install the frb repository.

Installing Dependencies

We have and will continue to keep the number of dependencies low. There are a few standard packages that must be installed and one package linetools under review for astropy affiliated status.

In general, we recommend that you use Anaconda for the majority of these installations.

Detailed installation instructions are presented below:

Python Dependencies

frb depends on the following list of Python packages.

We recommend that you use Anaconda to install and/or update these packages.

python versions 3.11 or later
numpy version 2.2 or later
astropy version 7.1 or later
scipy version 1.17 or later
healpy version 1.19 or later
pandas version 2.2 or later
requests version 2.18 or later
dust_extinction
matplotlib version 3.7 or greater
linetools version 0.3.1 or later
astropath version 0.1 or later

The following packages are required to access surveys (e.g. SDSS, DES) for data that may be associated to an FRB:

astroquery v0.4.11 or later
datalab-client v2.20 or later
pyvo version 0.9.2 or later

Extras

The following packages are more for builders and not recommended for most users.

The following package is required to map a slit onto a finder chart (frb.figures.finder):

photutils version 1.11.0 or later
scikit-image version 0.21.0 or later

The following are required to run some of the halo codes:

ne2001 NE2001
george :: Use pip
class version 2.7 or greater
Aemulator pip is broken..
hmf_emulator HMF emulator (requires GSL)

The following are required to build host galaxy objects:

pPXF version 6.7 or greater
pcigale version 2025.0

The following is required to run the code in dm_kde:

asymmetric_kde no versioning

The following is required to run the MCMC DM code in dm.mcmc.py:

numba version >= 0.50

For pPXF, you will also likely need to modify the standard install to use the Chabrier libraries. See the InstallNotes in this Google Drive.

The following are required for using functions in halos.photoz.py:

threedhst version >= 0.1.dev0
progressbar2 :: Use pip
pathos :: Use pip

Our CIGALE wrappers use custom filter files not provided by their current release (e.g DES, Pan-STARRS). See the instructions for adding those as needed.

Installing frb

Presently, you must download the code from github:

#go to the directory where you would like to install frb.
git clone https://github.com/FRBs/FRB.git

From there, you can build and install with:

cd FRB
python setup.py install  # or use develop

This should install the package and scripts. Make sure that your PATH includes the standard location for Python scripts (e.g. ~/anaconda/bin)

Adding additional CIGALE filter files

We have had to add new filters to CIGALE (e.g. from DES and Pan-STARRS). You can find these filter files in frb.data.analysis.CIGALE. If you use any of those surveys, our wrappers will not work without them.

Here are the steps to update CIGALE:

cd frb/data/analysis/CIGALE
Add any desired filter into your CIGALE code base with: pcigale-filters add file.dat

Note that DECaLs uses the BASS and MzLS data files.

EAZY setup

In order to perform photo-z estimation with EAZY using our wrappers, the following changes need to be made.

Add an environment variable EAZYDIR that points to your EAZY installation. Add this to your bashrc:
```
export EAZYDIR="/path/to/eazy-photoz/"
```
Locate the templates folder in $EAZYDIR and edit the paths present in *.spectra.param. Replace all SED file paths with the absolute paths. For instance, in $EAZYDIR/templates/eazy_v1.3.spectra.param, replace:
```
templates/EAZY_v1.1_lines/eazy_v1.1_sed1.dat
```

with:

/path/to/eazy-photoz/templates/EAZY_v1.1_lines/eazy_v1.1_sed1.dat

pPXF

Our pPXF wrapper currently uses an older version of the code (v 6.7.17) and a few custom files. Contact JXP if you wish to use this.