278261631/autophot

⚠️ Project on hold: I am currently on hiatus writing my PhD thesis - for queries with AutoPhOT please open an issues and I will address them as soon as possible - Seán

Introduction

AutoPhOT is novel automated pipeline that is designed for rapid, publication-quality photometry of transients.

The pipeline is built from the ground up using Python 3 - with no dependencies on legacy software. Capabilities of AutoPhOT include aperture and PSF-fitting photometry, template subtraction, and calculation of limiting magnitudes through artificial source injection. AutoPhOT is also capable of calibrating photometry against either survey catalogs (e.g. SDSS, PanSTARRS), or using a custom set of local photometric standards.

You can find the AutoPhOT paper here

Feedback is welcome. Email me at sean.brennan2@ucdconnect.ie

Installation

We suggest creating a new environment for AutoPhOT. This can be done using conda by running:

conda create -n autophot_env python=3.7

then to activate this environment, run:

conda activate autophot_env

Before installing autophot, you'll need conda-forge and astropy in your conda channels. See here for help.

Install AutoPhOT using the conda install:

conda install -c astro-sean autophot

If you want to update AutoPhOT, you can do so using:

conda update -c astro-sean autophot

Additional functionality

To fully utilise the AutoPhOT Code, several additional softwares may be used.

Astrometry.Net

AutoPhOT relies on Astrometry.net by Dustin Lang to solve for WCS. While the code can be downloaded/installed here and here we suggest using Homebrew to install Astometry.net.

brew install astrometry-net

To make sure everything is setup correctly, we can run the following in the terminal:

solve-field

In order for Astometry.net to run successfully, it requires pre-indexed files for calibration. Firstly, we can create a new directory called "data".

mkdir data
cd data/

Next, while in the data directory, we can run the following to obtain these index files (~40Gb):

wget -r -np http://broiler.astrometry.net/~dstn/4200/

This will download all the index files to the data/ folder. Once this download is completed, this data folder must be placed in the correct location so that Astrometry.net can find it.

We can search for the location of the solve-field command using the following:

which solve-field

It should be something similar to /usr/local/Cellar/astrometry-net/0.85_1/solve-field (although maybe not exactly). Move our data folder to the parent directory using:

cd ../
cp -R data /usr/local/Cellar/astrometry-net/0.85_1/

To update AutoPhOT on the location of Astrometry.Net, update (if needed) 'solve_field_exe_loc' in the autophot_input dictionary (see here for example).

If the user trusts their WCS this step can be ignored as Astrometry.net is not used.

HOTPANTS

AutoPhOT can use HOTPANTS by Andy Becker which can be found here.

We can download the HotPants code from Github using:

git clone https://github.com/acbecker/hotpants
cd hotpants\

Next we need to modify the Makefile. HOTPANTS requires CFITSIO to be already installed. We can install this using Homebrew:

brew install cfitsio

Which will install the library in a directory similar to /usr/local/Cellar/cfitsio/4.0.0 (although maybe not exactly). In this directory there should be two folders, include and bin.

We need to update the Makefile for HOTPANTS to work correctly. This file can be opened using a text editor. We need to update the CFITSIOINCDIR and LIBDIR variables to point towards the include and bin directories respectively.

CFITSIOINCDIR=/usr/local/Cellar/cfitsio/4.0.0/include
LIBDIR=/usr/local/Cellar/cfitsio/4.0.0/lib

Finally we can compile the code by running the following in the hotpants/ directory

make

Once installed, locate the hotpants executable and update 'hotpants_exe_loc' in autophot_input (see here for example) .

Known error with installation of HOTPANTS

There is a known bug with the HOTPANTS installation on MacOS - if upon installation you get 'malloc.h' file not found, replace:

#include <malloc.h>

with

#if !defined(  MACH  )
#include <malloc.h>
#endif
#if defined(  MACH  )
#include <stdlib.h>
#endif

to every .c file. Then you can run the make command.

If the user has no need for image subtraction or wants to use Zogy only, this step can be ignored.

ZOGY

AutoPhOT can also use Zogy which has a python wrapper and can be found here. We can install this straight from Github. Make sure the correct environment is activated. we can clone the Github repository to the current directory:

git clone https://github.com/dguevel/PyZOGY

and we can install it by running:

cd PyZOGY/
python setup.py install

no further action is required.

How to use AutoPhOT

I've designed AutoPhOT to be a user friendly pipeline, meaning you need very little prior knowledge on the art of photometry.

→ I've put together some Jupyter notebooks the get you started with AutoPhOT, click here.

→ If you wish to use the packages within AutoPhOT outside of the examples given, documentation on each package is given here.

→ A list of settings used in AutoPhOT can be found here.

→ If you need an example of how to use specific functions in AutoPhOT, please open an issue here.

Referencing & Attribution

If you use results from AutoPhOT in a publication, please cite Brennan & Fraser (2022). The AutoPhOT code is released under a GPL3 licence and you are free to reuse the code as you wish. If you modify AutoPhOT or use it in a strange fashion (or even if you use it normally), we make no guarantee that your photometry will be valid.

Testing and Debugging

If you experience errors with a particular file, the most effective means of debugging is to share the file with me (Seán Brennan) for diagnostic. Once bugs have been addressed all files will be deleted.

All shared data will be kept confidential.

278261631 / autophot