|CI|_ |Python|_ |License|_
.. |CI| image:: https://circleci.com/gh/tupui/otsensitivity.svg?style=svg .. _CI: https://circleci.com/gh/tupui/otsensitivity
.. |Python| image:: https://img.shields.io/badge/python-2.7,_3.7-blue.svg .. _Python: https://python.org
.. |License| image:: https://img.shields.io/badge/license-LGPL-blue.svg .. _License: https://opensource.org/licenses/LGPL
otSensitivity
What is it?
This project implements Sensitivity Analysis methods.
It is based on OpenTURNS <http://www.openturns.org>_.
The simplest analysis is to visualize how the quantity of interest respond individually to each input parameter:
.. code-block:: python
pairplot(sample, data)
.. image:: doc/images/ishigami_pairplot.png
This analysis can be completed by quantitative information. Variance-based analysis is commonly used:
.. code-block:: python
s, st = sobol_saltelli(function, 1000, 3, [[-np.pi, -np.pi, -np.pi],
[np.pi, np.pi, np.pi]])
plot_indices([s, st])
.. image:: doc/images/ishigami_indices.png
It is possible to use a polar coordinate system:
.. code-block:: python
plot_indices([s, st], polar=True)
.. image:: doc/images/ishigami_indices-polar.png
In case of an already existing sample, one can use density based measures:
.. code-block:: python
momi = moment_independent(X, Y)
delta = momi[2]['Delta']
plot_indices([delta])
.. image:: doc/images/ishigami_moment.png
This method use not only the variance but all the PDF in order to compute sensitivity information. Also, it does not require the use of any particlar sampling design.
How to install?
Requirements ............
The dependencies are:
- Python >= 2.7 or >= 3.3
-
numpy <http://www.numpy.org>_ >= 0.10 -
scipy <http://scipy.org>_ >= 0.15 -
OpenTURNS <http://www.openturns.org>_ >= 1.12 -
matplotlib <https://matplotlib.org>_ >= 1.5.3
Installation ............
Using the latest python version is prefered! Then to install::
git clone [email protected]:.../otsensitivity.git
cd otsensitivity
python setup.py install