Hi Mattheo, 
I totally agree with that feature request and am volunteer to work on that
if needed.

I have been doing experiments with processing scripts using library such as
pandas (relying on matplotlib), seaborn and plot.ly.  
See "Summarize_numeric_columns_using_Pandas" script in  my github
https://github.com/aeag/QGIS-Processing/tree/master/scripts ) for a first
example. We also have unpublished scripts making graphs if needed.

I expect to have a student working on that this year: exploring what libs
are worth being shipped with qgis, and developping some processing tools
using that. 


The tries I have made up to now seems to conclude that the "must have" lib
is Pandas. Many QGIS processing tools working on non geographic data are
limited when datasets grow up. Pandas let it be extrmely fast and concise,
in a syntax very close from "R". That may be worth shipping it.. 
It weights nothing in comparison of Numpy (already required).. 

Once pandas shipped, we could probably make great improvements to Statist
and its friends. 

Cheers
Régis



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/graph-and-plots-improvements-tp5194130p5194168.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
_______________________________________________
Qgis-developer mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Reply via email to