Tleco
stands for both in the fire and rise in the nahuatl language. Tleco
is a numerical code that simulates particles in relativistic plasma, and the rise of radiation from accelerating particles.
Tleco has already been used for several publications (previously called Paramo):
- Rueda-Becerril, J. M., Harrison, A. O., & Giannios, D. 2021, MNRAS, 501, 4092, doi: 10.1093/mnras/staa3925
- Davis, Z., Rueda-Becerril, J. M., & Giannios, D. 2022, MNRAS, 513, 5766, doi: 10.1093/mnras/stac1282
- Combi, L., & Siegel, D. M. 2023, ApJ, 944, 28, doi: 10.3847/1538-4357/acac29
Ensure you have Python 3.9 or later installed on your system. Tleco can be installed via pip and works on all major operating systems. Using and building with Tleco does not require any compiler. A rust compiler is necessary to develop Tleco.
pip install tleco
Tleco consist of both Rust functions and Python functions. Follow the example below to get access to correct functions:
import tleco #tleco.rust_functions
from tleco import constants as cons #from tleco gives access to the python files and functions
Getting started with Tleco is straightforward and several examples can found in Examples]. Follow these instructions to run the example 'broken_pwl_law_full_example.py'.
git clone https://github.com/zkdavis/Tleco.git
cd Tleco
pip install -r requirements.txt
pip install tleco
python examples/broken_pwl_law_full_example.py
Contributors and developers are welcome to enhance Tleco's features or address issues. Following the steps above for running the broken power-law example will get you nearly ready for development. You will still need to install a rust compiler(https://www.rust-lang.org/tools/install). Once installed if any of the rust files are changed make sure to rebuild the object with maturin:
maturin develop
- Fork the repository.
- Create your feature branch (git checkout -b feature/AmazingFeature).
- Commit your changes (git commit -am 'Add some AmazingFeature').
- Push to the branch (git push origin feature/AmazingFeature).
- Open a pull request.
This section is not complete and will be updated over time
-
ic_iso_monochrome - computes the emissivity(
$\frac{ergs}{cm^3 Sr}$ ) at frequency nuout(Hz) from inverse Compton (IC) scattering in an isotropic photon field assuming the photonfield is monochromatic [see in src/radiation.rs:L356]-
Parameters:
-
nuout
(f64): frequency(Hz) in the comoving frame to compute emission at. -
uext
(f64): energy density($\frac{ergs}{cm^-3}$ ) of the external photon field in the comoving frame. -
nuext
(f64): frequency(Hz) in the comoving frame of the external photon field. -
n
(&Array1): particle distribution as function of lorentz factor -
g
(&Array1): Lorentz factor grid
-
-
Returns:
-
jnu
(f64): emissivity($\frac{ergs}{cm^3 Sr}$ ) for frequency nuout
-
-
Parameters:
-
ic_iso_monochrome_full - computes the emissivity(
$\frac{ergs}{cm^3 Sr}$ ) from inverse Compton (IC) scattering in an isotropic photon field assuming the photonfield is monochromatic [see in src/radiation.rs:L406]-
Parameters:
-
freqs
(&Array1): frequency(Hz) array in the comoving frame to compute emission over. -
uext
(f64): energy density($\frac{ergs}{cm^-3}$ ) of the external photon field in the comoving frame. -
nuext
(f64): frequency(Hz) in the comoving frame of the external photon field. -
n
(&Array1): particle distribution as function of lorentz factor -
g
(&Array1): Lorentz factor grid
-
-
Returns:
-
jic
(Array1): emissivity($\frac{ergs}{cm^3 Sr}$ ) for frequency range freq
-
-
Parameters:
-
rad_cool_pwl - computes the radiative inverse Compton cooling (
$\frac{\partial g}{\partial t}$ [$s^{-1}$]) from isotropic photon field uu($\frac{ergs}{cm^-3}$ ) [see in src/radiation.rs:L502]-
Parameters:
-
gg
(&Array1): Lorentz factor grid -
freqs
(&Array1): frequency(Hz) array in the comoving frame. -
uu
(&Array1): energy density($\frac{ergs}{cm^-3}$ ) of photon field in the comoving frame for every frequency in freqs. -
with_kn
(bool): bool that will include Klein Nishina affects to the cross section when true.
-
-
Returns:
-
dotg
(Array1): radiative cooling ($\frac{\partial g}{\partial t}$ [$s^{-1}$])
-
-
Parameters:
-
rad_cool_mono - computes the radiative inverse Compton cooling (
$\frac{\partial g}{\partial t}$ [$s^{-1}$]) from isotropic monotonic photon field u0($\frac{ergs}{cm^-3}$ ) [see in src/radiation.rs:L557]-
Parameters:
-
gg
(&Array1): Lorentz factor grid -
nu0
(f64): frequency(Hz) in the comoving frame of the photon field u0. -
u0
(f64): energy density($\frac{ergs}{cm^-3}$ ) of photon field in the comoving frame. -
with_kn
(bool): bool that will include Klein Nishina affects to the cross section when true.
-
-
Returns:
-
dotg
(Array1): radiative cooling ($\frac{\partial g}{\partial t}$ [$s^{-1}$])
-
-
Parameters:
- power_law - Computes a power law function. [see in pysrc/tleco/misc_func.py:L33]
- Parameters:
- Returns:
f
(**): Array of output values according to the power law.f = np.zeros_like(g)bounds = (g >= g_min) & (g <= g_max)f[bounds] = np.power(g, p)[bounds]if(normalize):f = f/np.trapz(f,g)return f*n0
The following Python packages are used in this project:
- astropy 6.0.1
- ebltable 0.5.2
- matplotlib 3.8.3
- numpy 1.26.4
- pytest 8.1.1
- scipy 1.13.0
- toml 0.10.2
- tqdm 4.66.2
- maturin 1.5
This project requires Python ^3.9.