topological-data-analysis

You compute persistent homology and Betti numbers by applying this Skill to point clouds, which calculates topological features across a filtration and summarizes them using persistence diagrams, barcodes, and Betti numbers.

To convert persistence diagrams into features for machine learning pipelines, use topology-aware feature engineering to transform persistence outputs into vector representations like persistence landscapes, Betti curves, and persistence images.

Yes, you can extract topological features from image data using cubical complexes. This Skill supports processing image data to compute persistent homology and identify underlying physical or biological structures.

Yes, ripser works within this Skill's environment alongside scikit-learn, utilizing libraries like persim to compute Wasserstein or bottleneck distances to compare persistence diagrams for downstream modeling.

You need Mapper to build a topological network from a point cloud when you want to visualize and extract multi-scale shape structures, such as loops and connected components, that standard distance-based methods often miss.

To apply TDA feature engineering to time-series embeddings for neuroscience data, use this Skill to compute persistent homology from the embeddings and convert the resulting persistence diagrams into vectorized representations for downstream modeling.