h-xie / ztf_sim

:telescope: Zwicky Transient Facility survey scheduler

ztf_sim

🔭 Scheduling library for the Zwicky Transient Facility.

Implements the Integer Linear Programming scheduling algorithm described in Bellm et al. 2019 (PASP 131, 1000).

Installation

You will need a license for the Gurobi optimizer on the machine you want to run simulations on. Free academic licenses are readily available.

You are strongly encouraged to use conda and a conda environment for the installation.

conda create -n ztf_sim_test

conda activate ztf_sim_test

conda install python=3.7


conda config --add channels conda-forge 
conda config --add channels http://conda.anaconda.org/gurobi
conda install pip numpy scipy astropy astroplan pandas scikit-learn xgboost sqlalchemy gurobi


pip install sklearn_pandas 
pip install transitions

pip install -e git+https://github.com/ZwickyTransientFacility/ztf_sim.git#egg=ztf_sim

(To remove the environment, use conda remove --name ztf_sim_test --all.)

Configuration

Scheduler Configuration

The scheduler configuration determines which observing programs will run (fields, filters, cadences, etc.)

An example set of scheduler configuration file is provided in sims/.

You can copy them locally with

wget https://raw.githubusercontent.com/ZwickyTransientFacility/ztf_sim/master/sims/example_scheduler_config.json
wget https://raw.githubusercontent.com/ZwickyTransientFacility/ztf_sim/master/sims/survey_180501.json
wget https://raw.githubusercontent.com/ZwickyTransientFacility/ztf_sim/master/sims/reference_building.json

Simulation Configuration

The simulation configuration determines which nights to simulate, which historical weather to use, and whether to overwrite the existing simulated database.

An example configuration file is provided in config/default.cfg.

You can copy it locally with wget https://raw.githubusercontent.com/ZwickyTransientFacility/ztf_sim/master/config/default.cfg

Running

run_ztf_sim --help summarizes the argument of the command line driver for the simulations. Assuming you've copied the configuration files to your current directory, you should now be able to run

run_ztf_sim example_scheduler_config.json default.cfg

which will write an sqlite database file named example_ztf_schedule.db to the current directory.

Output

The simulated schedule is written to a SQLite database in the LSST Operations Simulator format with a few additional columns. Example code for reading and summarizing the simulated schedule is located in the bin/ directory, and can be run as

analyze_ztf_sim example_ztf_schedule.db

The simulator also writes a logfile, which may be monitored while the simulation is running with tail -f example_ztf_schedule_log.txt