Skip to content

andreatramacere/jetset

Repository files navigation

pip build-test conda build-test

img

JetSeT is an open source C/Python framework to reproduce radiative and accelerative processes acting in relativistic jets,
allowing to fit the numerical models to observed data. The main features of this framework are:

  • handling observed data: re-binning, definition of data sets, bindings to astropy tables and quantities definition of complex numerical radiative scenarios: Synchrotron Self-Compton (SSC), external Compton (EC) and EC against the CMB

  • Constraining of the model in the pre-fitting stage, based on accurate and already published phenomenological trends. In particular, starting from phenomenological parameters, such as spectral indices, peak fluxes and frequencies, and spectral curvatures, that the code evaluates automatically, the pre-fitting algorithm is able to provide a good starting model,following the phenomenological trends that I have implemented. fitting of multiwavelength SEDs using
    both frequentist approach (iminuit) and bayesian MCMC sampling (emcee)

  • Self-consistent temporal evolution of the plasma under the effect of radiative, accelerative processes, and adiabatic expansion. Both first order and second order (stochastic acceleration) processes are implemented.

Acknowledgements

If you use this code in any kind of scientific publication please cite the following papers:

Licence

JetSeT is released under a 3-clause BSD license, for deatils see License file

Documentation

visit: https://jetset.readthedocs.io/en/latest/

run the notebook on binder: Binder

Installation

NOTE: Starting from version 1.1.0, python 2 is not supported anymore. Python >=3.8 is suggested, older python 3 versions (< 3.8) should work.

Install JetSeT from Anaconda

  • create a virtual environment (not necessary, but suggested):

    conda create --name jetset python=3.9 ipython jupyter

    conda activate jetset

  • install the code:

    conda install -c andreatramacere jetset

    if conda fails with dependencies you can try

    conda install -c andreatramacere -c astropy -c conda-forge 'jetset>=1.2'

Install JetSeT from pip

  • MacOS

    • create a virtual environment (not necessary, but suggested): pip install jetset
  • Linux

    NOTE: currently, pip binaries for linux are not provided, so pip will build the binary on the fly. Hence:

    • install swig with of these methods and check that you have a C compiler (gcc)

    • pip install jetset

    if fails, use one of the following methods

    1. Use anaconda

    OR

    1. Install from source

    OR

    1. Install the git release binaries from git releases

Install JetSeT GitHub binaries

Install the git release binaries from git releases

run the test

pytest --disable-warnings  --pyargs  -vvv jetset.tests.test_users::TestUser

Install the JetSeT from source

Download the code

Installation from source using Anaconda

  • Install requirements, run on the command line:
    • conda install --yes swig">=3.0.0"
    • conda install -c astropy -c conda-forge --file requirements.txt

if anaconda fails to install swig, you can try one of the following alternative methods

  • run on the command line:

    • python setup.py clean
    • python setup.py install
  • run the test (optional, run all the examples outside the installation dir)

cd ~/
mkdir test_jetset
cd test_jetset
pytest  --pyargs  -vvv jetset.tests.test_users::TestUser

Installation from source using PIP

  • Install requirements, run on the command line:

    • pip install swig>=3.0.0
    • pip install -r requirements.txt

    if pip fails to install swig, you can try one of the following alternative methods

  • run on the command line:

    • python setup.py clean
    • python setup.py install
  • run the test (optional, run all the examples outside the installation dir)

cd ~/
mkdir test_jetset
cd test_jetset
pytest  --pyargs  -vvv jetset.tests.test_users::TestUser

Requirements

The following python packages are required:

    python =3.9 (python 3.9 is suggested, older python 3 versions should  work, python 2 is not supported any more from version>=1.1.0)
    setuptools
    scipy
    numpy
    astropy
    matplotlib
    swig
    future
    iminuit
    corner
    six
    emcee
    pyyaml
    numba
    sherpa

A C compiler is also necessary, plus the SWIG wrapper generator.

All the dependencies are installed following the Anaconda method OR the pip method, as described above.

jetset code repoistory

The code is hosted here: