Triangular Distribution Plotting (aka corner plots) for MCMC Sampling Analysis.
This is a very early release of the code, so please create an issue or submit a pull request.
- Dynamic and Fancy Plotting
- Includes KDE (Kernel Distribution Estimation) for 2-D contours
The the most recent and stable version can be installed manually, by cloning the repository:
$ git clone https://github.com/Relativist1/Tardis.git
$ cd Tardis
$ python3 setup.py install# samples are either imported or directly used after mcmc sampler
from tardis import Tardis
m_true = -0.9594
b_true = 4.294
f_true = 0.534
truths = [m_true, b_true, f_true]
labels = [r"$m_{true}$", r"$b_{true}$",r"$f_{true}$"]
# if Emcee ensemble.sampler is used
samples = sampler.get_chain(flat=True)
Tardis(samples, truths=truths, labels =labels,
savefig='new1.png', diag_shade_color='red',
shade=True, truth1d=True, truth2d = False)
The package is constantly under development.
Coming soon....
Tardis is licensed under GPLv3. See LICENSE for more details.