gtda.mapper: Mapper




Projection onto specified columns.

mapper.Eccentricity([exponent, metric, …])

Eccentricities of points in a point cloud or abstract metric space.


Entropy of rows in a two-dimensional array.


mapper.OneDimensionalCover([kind, …])

Cover of one-dimensional data coming from open overlapping intervals.

mapper.CubicalCover([kind, n_intervals, …])

Cover of multi-dimensional data coming from overlapping hypercubes (technically, parallelopipeds) given by taking products of one-dimensional intervals.


mapper.FirstSimpleGap([linkage, affinity, …])

Agglomerative clustering cutting the dendrogram at the first instance of a sufficiently large gap.

mapper.FirstHistogramGap([linkage, …])

Agglomerative clustering with stopping rule given by a histogram-based version of the first gap method, introduced in [1]_.

mapper.ParallelClustering(clusterer[, …])

Employ joblib parallelism to cluster different portions of a dataset.

Nerve (graph construction)

mapper.Nerve([min_intersection, …])

1-skeleton of the nerve of a refined Mapper cover, i.e.


mapper.make_mapper_pipeline([scaler, …])

Construct a MapperPipeline object according to the specified Mapper steps [1]_.


Subclass of sklearn.pipeline.Pipeline to deal with pipelines generated by make_mapper_pipeline.


mapper.plot_static_mapper_graph(pipeline, data)

Plot Mapper graphs without interactivity on pipeline parameters.

mapper.plot_interactive_mapper_graph(…[, …])

As of version 0.5.0, we recommend using the object-oriented interface provided by :class:`MapperInteractivePlotter` instead of this function.

mapper.MapperInteractivePlotter(pipeline, data)

Plot Mapper graphs in a Jupyter session, with interactivity on pipeline parameters.


mapper.method_to_transform(cls, method_name)

Wrap a class to add a transform method as an alias to an existing method.


Construct a transformer from a callable acting on 1D arrays.