maemo.org - Talk - [Announce] (Calculation with) Uncertainties for Maemo 4 & 5 & MeeGo Harmattan & Sailfish

Uncertainties is Python module for calculations with uncertainties, that transparently handles calculations with numbers with uncertainties (i.e., error propagation). It can also yield the derivatives of any expression.

Whatever the complexity of the calculation, this package returns the result with its uncertainty as predicted by linear error propagation theory. It automatically calculates derivatives and uses them for calculating uncertainties. Correlations between variables are thus automatically handled, which sets it apart from many existing error propagation codes.

Mathematically, numbers with uncertainties are, in this package, probability distributions. They are not restricted to normal (Gaussian) distributions and can be any kind of distribution. These probability distributions are reduced to two numbers: a nominal value and a standard deviation.

So, the Uncertainties package calculations are more refined than the simplified error propagation calculations commonly used in laboratories.

Uncertainties 2.4.1 is released for Maemo 4 Diablo, Maemo 5 Fremantle. And Uncertainties 2.4.6 is released for MeeGo 1.2 Harmattan and Mer / Nemo / Sailfish OS.

Screenshot of a Nokia N900 running Uncertainties 2.4.1 inside IPython 0.10.2 :
http://www.RobertoColistete.net/Pyth...ies_N900_2.png

Screenshot of a Nexus 4 with Sailfish OS running Uncertainties 2.4.6 (within IPython 2.0.0) :
https://openrepos.net/sites/default/...sailfishos.jpg

"Uncertainties" history for Maemo/MeeGo/Sailfish :
- version 1.8 for Maemo 4 and Maemo 5 (22/02/2012);
- version 1.8 for MeeGo Harmattan (01/08/2012);
- version 1.9 for Maemo 4, Maemo 5 and MeeGo Harmattan (31/12/2012), then Mer (19/02/2013);
- version 2.3.6 for Maemo 4, Maemo 5, MeeGo Harmattan and Mer (05/06/2013);
- version 2.4.1 for Maemo 4, Maemo 5, MeeGo Harmattan and Mer (05/08/2013).
- version 2.4.6 for Mer / Nemo / Sailfish OS (04/05/2014).
- version 2.4.6.1 for MeeGo Harmattan (23/09/2014).

See here the version history of the "Uncertainties" project.

Install & use on Maemo 4 & 5 :
Enable the extras repository on Maemo 4, and the extras-testing or extras-devel repository on Maemo 5. Then, in X Terminal, as root :
# apt-get update
# apt-get install python-uncertainties
Uncertainties uses just 0.3 MB after install and is optified (on Maemo 5).
If you want to deal with arrays of numbers with uncertainties, it is possible to use Uncertainties with Numpy (not available on Maemo 4), so to install it :
# apt-get install python-numpy

Install & use on MeeGo Harmattan :
Two options of repositories :
1) configure my Harmattan repository;
or
2) install Warehouse (OpenRepos.net client) for MeeGo Harmattan, then open Warehouse, search for "Uncertainties", and enable the repository.
After configuring one of the above repository options, open the Terminal, type :
$ devel-su
and the 'rootme' password, then
# apt-get install python-uncertainties
If you want to deal with arrays of numbers with uncertainties, it is possible to use Uncertainties with Numpy, so to install it :
# apt-get install python-numpy

Install & use on Sailfish OS :
Install Warehouse (OpenRepos.net client) for Sailfish OS, see the Jolla Tides guide with screenshots.
Then open Warehouse, search for "Uncertainties", enable the repository, then install it.
If you want to deal with arrays of numbers with uncertainties, it is possible to use Uncertainties with Numpy. To install it, search for "NumPy" in Warehouse and install it.

To use interactively, IPython is highly recommended, for example :
In [1]: import uncertainties as un
To show a long help :
In [2]: ?un
To show ufloat help :
In [3]: ?un.ufloat
See the screenshot above and the Uncertainties site documentation for more examples.

For Maemo 5 users, if you want Uncertainties promoted to extras repository, please vote in :
Uncertainties 2.4.1 Maemo1

I am planning a simple GUI using PyQt / PySide & QML for Maemo 4 & 5, MeeGo Harmattan and Nemo/Sailfish OS when I have enough spare time.