Skip to content

cositools/cosi-data-challenges

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

COSI Data Challenge

Required Software

The data challenge module requires the MEGAlib code, available here. Among other things, MEGAlib simulates the emission from any (MeV) gamma-ray source, simulates the instrument response, performs the event reconstruction, and performs the high-level data analysis. See the above link for more details regarding the MEGAlib package.

Getting Help

For any help/problems with running the data challenge module please contact Chris Karwin at: [email protected].

Data Challenge Releases

  • March 2023: The first data challenge is now available! It can be found here.
  • March 2024: The second data challenge is now available! It can be found here.

Purpose

The main purpose of this repository is to simulate the all-sky data that will be observed by COSI. The primary code is run_dc.py, which can be called with run_sims.py, with the main input parameters passed via inputs.yaml. Additionally, parallel simulations with multiple time bins can be ran using run_parallel_sims.py, which distributes the time bins to seperate compute nodes. The pipeline also supports the use of mcosima with numerous cores per compute node. The modules can be ran directly from the command line, or submitted to a batch system, which allows them to be easily employed for generating multiple/long simulations.

Available Sources for Simulations

See Source_Library for available sources. Let us know if you want any specific source added!

Quickstart Guide

1. Download cosi-data-challenges directory:
  $ git clone https://github.com/cositools/cosi-data-challenges.git

2. Install with pip:
  $ cd cosi-data-challenges
  $ pip install -e .

3. Start a new analysis directory, and enter the commmand-line prompt:
  $ make_sim
   
4. Specify inputs in inputs.yaml 
     
5. Run setup script: 
  $ python run_sim_setup.py
  - This will setup the source directory and copy all needed files for running the code.
  
6. To run the code:  
  - Uncomment the functions inside run_sims.py that you want to run.
  - The code can be ran directly from the terminal or submitted to a batch system.
  - The code supports both PBS and Slurm.
  - To run from the terminal use python run_sims.py.
  - To run parallel jobs in cosima with numerous time bins use python run_parallel_sims.py. 
  - To submit a single job use 'python submit_jobs.py' for PBS and 'sbatch slurm_single.sh' for Slurm. 

7. If running parallel jobs:
  - Run: python run_parallel_sims.py.  
  - This will setup all the sim directories and the scripts needed for running parallel jobs, depending on the run type specified in inputs.yaml. 

8. Note that the batch submission commands may need to be modified based on the user's specific batch system and needs.
  - The batch system is specified via the run_type parameter. 
  - The example directory contains different batch scripts. 

About

Simulation pipeline for the COSI data challenges

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •