pathpyG
pathpyG copied to clipboard
GPU-accelerated Next-Generation Network Analytics and Graph Learning for Time Series Data on Complex Networks.
I just saw that `PyG` is planning to integrate all time-related stuff into `Data`, thus, deprecating `TemporalData`. We need to modify our `TemporalGraph` accordingly. (see https://github.com/pyg-team/pytorch_geometric/issues/3230) BTW: This is also...
Numpy released the version 2.0 recently. This currently leads to the following problem: ``` A module that was compiled using NumPy 1.x cannot be run in NumPy 2.0.0 as it...
As far as I am aware, we reused the PathPy3 code for the implementation of the random walks. I noticed the following problems that need to be fixed: - [x]...
Implemented new netzschleuder support, which addresses the following issues: - [x] simplify implementation by downloading CSV data that are parsed based on new pandas interface - [x] improve robustness of...
Turning a directed graph with isolated nodes into an undirected graph yields an error: Minimal example ``` import pathpyG as pp g = pp.Graph.from_edge_list([('a', 'b')], num_nodes=3) g_u = g.to_undirected() ```...
Replace code that parses graphtool binary format by code that downloads network in csv-based format - [ ] move graphtool parser to optional function or separate module in io -...
The task would be to implement a set of basic community detection algorithms, such as: - [ ] [InfoMap](https://www.mapequation.org/assets/publications/RosvallBergstromPNAS2008Full.pdf) - [ ] [Louvain method](https://iopscience.iop.org/article/10.1088/1742-5468/2008/10/P10008) - [ ] [Modularity Optimization](https://www.pnas.org/doi/full/10.1073/pnas.0601602103) -...
The task would be to implement the higher-order force-directed layout algorithm from the following paper: Perri, V., Scholtes, I. (2020). HOTVis: Higher-Order Time-Aware Visualisation of Dynamic Graphs. In: Auber, D.,...
The task would be to implement, test, and evaluate the Bayesian model selection procedure described in the following paper: Luka V. Petrovic and Ingo Scholtes. 2022. Learning the Markov Order...
The model selection fails with an IndexError for the London Tube data set. Minimal code to reproduce the error: ``` import pathpyG as pp paths_tube = pp.PathData.from_ngram('../data/tube_paths_train.ngram', sep=',', weight=True) m...