This project develops a simple, fast and easy to use Python graph library using NumPy, Scipy and PySparse. For information on installing, and usage, see http://packages.python.org/apgl/index.html.
UPDATE: This project is no longer maintained. I recommend networkx (https://networkx.github.io/) or igraph (http://igraph.org/python/) instead.
Changes in version 0.8.1:
- addVertices method.
- Documentation for CsArrayGraph
- Read properties from config file
Changes in version 0.8:
- Added CsArrayGraph which is based on sppy
- General tidy up: removed feature extraction and prediction code which is now at https://github.com/charanpald/sandbox
- Several bug fixes and optimisations
Changes in version 0.7.3:
- DictGraph - toSparseGraph, added depth and breadth first search, degree sequence, num directed edges, dijkstra's algorithm, adjacency list, and find all distances.
- Matrix graphs - toDictGraph, fixed depth and added breadth first searches
- SparseGraph - toCsc, toCsr, specify sparse matrix format in constructor
- Other minor changes
Changes in version 0.7.2:
- Constructors for graph classes accept a size and weight matrix.
- len property for vertex lists
- Optimisation of depth first search, and ABCSMC
- Lots of bug fixes and other improvements
Changes in version 0.7.1:
- Pickling of graphs
- Update of ABCSMC algorithm for better multiprocessing
Changes in version 0.7:
- Python 3 support
- Windows installer
- DictTree - getRootId, children, leaves, nonLeaf, copy, pruning and subtree methods
- Sampling returns tuple of ndarrays
- Other minor changes and documentation improvements
This library was written by Charanpal Dhanjal (email: charanpal at gmail dot com). If you find it useful, please cite APGL as follows:
Charanpal Dhanjal, An Introduction to APGL, Statistics and Applications Group, Telecom ParisTech, 2012