Show HN: Pysimplicial, Python library for simplicial complexes in topological ML (github.com)

The newly announced Python library, **Pysimplicial**, is an experimental toolkit aimed at facilitating simplicial triangulations and topological machine learning experiments. Developed from research focusing on Open-Closed State-sum Neural Networks, this library is tailored for researchers interested in topological deep learning rather than production applications. Pysimplicial offers functionalities like visualizing triangulations and calculating topological invariants, with built-in converters for adapting structures for use in various machine learning frameworks such as Graph Neural Networks (GNNs) and Multi-Layer Perceptrons (MLPs). Significantly, Pysimplicial's emphasis on topological invariants—like the genus, connected components, and the application of Pachner moves—opens new avenues for integrating topology into machine learning paradigms. This integration could enhance model robustness and data representation, particularly in high-dimensional spaces where traditional methods may struggle. As the toolkit is early in its development stage, it invites contributions from the community, providing a collaborative platform for advancing topological concepts in machine learning and expanding the potential of neural networks within this innovative field.