gtda.diagrams: Persistence diagrams

The module gtda.diagrams implements transformers to preprocess persistence diagrams, extract features from them, or compute pairwise distances between diagrams.



Replaces all homology dimensions in persistence diagrams with numpy.inf.

diagrams.Scaler([metric, metric_params, …])

Linear scaling of persistence diagrams.

diagrams.Filtering([homology_dimensions, …])

Filtering of persistence diagrams.


diagrams.PairwiseDistance([metric, …])

Distances between pairs of persistence diagrams.


diagrams.PersistenceLandscape([n_layers, …])

Persistence landscapes of persistence diagrams.

diagrams.BettiCurve([n_bins, n_jobs])

Betti curves of persistence diagrams.

diagrams.HeatKernel([sigma, n_bins, n_jobs])

Convolution of persistence diagrams with a Gaussian kernel.

diagrams.PersistenceImage([sigma, n_bins, …])

Persistence images of persistence diagrams.

diagrams.Silhouette([power, n_bins, n_jobs])

Power-weighted silhouettes of persistence diagrams.


diagrams.Amplitude([metric, metric_params, …])

Amplitudes of persistence diagrams.

diagrams.PersistenceEntropy([normalize, …])

Persistence entropies of persistence diagrams.


Number of off-diagonal points in persistence diagrams, per homology dimension.

diagrams.ComplexPolynomial([n_coefficients, …])

Coefficients of complex polynomials whose roots are obtained from points in persistence diagrams.