NNGT

Unified generation and analysis of networks in Python, with neuroscientific additions

Downloads
1.1K
Stars
14
Committers
3

Bot releases are visible (Hide)

NNGT - Bugfix for `connect_neural_{groups/types}`

Published by Silmathoron about 7 years ago

Corrected a wrong test that made groups only able to be either sources or targets but not both at the same time.

NNGT - Major internal improvements, upgraded doc and examples

Published by Silmathoron about 7 years ago

A lot of internal improvements have been added and the basic functions have been more thoroughly tested.

  • Improved plotting tools and analyses (added automatic binning + basic network drawing)
  • Better attributes handling
  • Additional correction to networkx, igraph and graph-tool handling
  • Shape is now saved in IO
  • Using logging
NNGT - Version 0.6: partial support for Shapely and multithreading

Published by Silmathoron over 7 years ago

With this new release, three main updates have been added:

  • support for custom network spatial structures through the geometry module, from the SENeC package PyNCulture. So far, edge distance is still computed as the simple distance between neurons, not taking potential obstacles into account.
  • multithreaded algorithms for fixed/gaussian_degree methods, as well as the distance_rule generator.
  • set_minis method to simulate synaptic noise from spontaneous release of neurotransmitters.

Other improvements:

  • added logging
  • removed globals.py and moved features to lib
  • improved support for edge attributes (still a work in progress)
  • custom graph plotting with matplotlib (work inprogress)

Planned improvements for next release:

  • support of all main graph formats for IO
  • full support of node and edge attributes
  • more complete tests for Travis
NNGT - Version 0.5: support of python 2/3

Published by Silmathoron over 8 years ago

Further optimization of library support with direct inheritance of library graph objects by Graph and its subclasses (this means that networkx, igraph and graph-tool functions can now be used directly on Graph instances).
The library now fully supports all python versions from 2.7 to 3.5 and continuous integration through Travis is on its way.
New graph generators (fixed_degree, distance_rule) have also been added, as well as analysis functions to characterize the activity of the networks modeled with NEST.
Eventually, basic graph IO has been implemented.

Planned improvements for next release:

  • support of all main graph formats for IO
  • full support of node and edge attributes
  • more complete tests for Travis
NNGT - Version 0.4: speed optimization and full multi-library support

Published by Silmathoron almost 9 years ago

Here comes the 2nd release for my favorite debuggers ;)

NNGT - Third version of development stage

Published by Silmathoron almost 9 years ago

This is a first step for the NNGT package: it's going to be tested at that point, so though it's not ready yet, let's call it the first official non-released version... yeah!

Package Rankings
Top 13.27% on Pypi.org
Related Projects