pathpyG
pathpyG copied to clipboard
GPU-accelerated Next-Generation Network Analytics and Graph Learning for Time Series Data on Complex Networks.
While fixing an issue with the `Graph.weighted_outdegrees()`-function (#253), I found some inconsistencies that we should probably be addressed: - Why is there a function for computing the weighted outdegrees, but...
Wehn importing the latest version, torch issues the following warning: ``` /opt/conda/lib/python3.10/site-packages/torch/cuda/__init__.py:138: UserWarning: CUDA initialization: Unexpected error from cudaGetDeviceCount(). Did you run some cuda functions before calling NumCudaDevices() that might...
This PR adds the `TemporalPercolation` implementation to pathpyG. ### Key Features: Temporal Graph Processing: Constructs event graphs from temporal edge lists using pathpyG. Connected Components Analysis: Identifies connected components in...
It would be great to have a more streamlined and user-friendly method for loading networks directly from an SQL database. Currently, the process works using the following code: ```python conn...
The current approach of calculating the number of directed edges in an undirected network is confusing and may lead to misinterpretation. For algorithms involving undirected networks, users are required to...
As discussed you find the method with a smaller memory footprint in this PR.
If `n==m`, `data.sort_by_time()` does not seem to sort the edges properly. My guess is that it instead sorts the node attributed then. As a temporary fix, I added a manual...
Creating the 3rd-order Model using the ["Kenyan Households"](http://www.sociopatterns.org/datasets/kenyan-households-contact-network/) dataset fails because the dataset seems to be too large. But the number of first-order edges is only 65 286 which seems...
Implement the following enhancements: - [x] Use `.toml` - [ ] Enable local config
Adding graphs currently only works for first-order graphs because of the non-hashable np.ndarray-node_ids in the mapping for higher-order graphs.