fds 0.0.3.dev1

Sensitivity analysis of chaotic simulations

.. figure:: https://travis-ci.org/qiqi/fds.svg?branch=master
:alt: Travis CI

.. toctree::
:maxdepth: 5

tutorials/src/vanderpol_python/vanderpol


What's it for
~~~~~~~~~~~~~

fds is a research tool for computational simulations that exhibis
chaotic dynamics. It computes sensitivity derivatives of time averaged
quantities, a.k.a. statistics, with respect simulation parameters.

For an introduction of chaotic dynamics, I highly recommend `Strogatz's
excellent book <https://www.amazon.com/gp/product/0813349109>`__. For a
statistical view of chaotic dynamical systems, please refer to
`Berlinger's
article <http://www.uvm.edu/~pdodds/files/papers/others/1992/berliner1992a.pdf>`__
Algorithm used in this software is described in `the upcoming AIAA
paper <https://dl.dropbox.com/s/2e9jxjmwh375i01/fds.pdf>`__

Download and use
~~~~~~~~~~~~~~~~

The best way to download fds is using pip. Pip is likely included in
your Python installation. If not, see `instruction
here <https://pip.pypa.io/en/stable/installing/>`__. To install fds
using pip, simply type

::

sudo pip install fds

Tutorials
~~~~~~~~~

- `First example -- Van der Pol oscillator in Python <tutorials/src/vanderpol_python/vanderpol>`__
- `Lorenz attractor in C <docs/tutorials/lorenz_c.md>`__
- `Lorenz 96 in MPI and C <docs/tutorials/lorenz96_mpi.md>`__

Guides
~~~~~~

- `Chaos and statistical convergence <docs/guides/statistics.md>`__
- `Lyapunov exponents and time
segmentation <docs/guides/lyapunov.md>`__
- `Save and restart <docs/guides/save_restart.md>`__

Reference
~~~~~~~~~

- `The least squares shadowing algorithm <docs/ref/lss_algorithm.md>`__
- `Function reference <docs/ref/function_ref.md>`__
- `License <LICENSE.md>`__