|
| 1 | +""" |
| 2 | +This library provides a Python implementation of the TDA Mapper algorithm, |
| 3 | +which is used for topological data analysis (TDA). The TDA Mapper algorithm |
| 4 | +is a method for extracting topological features from high-dimensional data |
| 5 | +by constructing a simplicial complex that captures the shape of the data. |
| 6 | +
|
| 7 | +The `tdamapper` package includes the following main modules: |
| 8 | +- `core`: Contains the core implementation of the Mapper algorithm. |
| 9 | +- `cover`: Provides classes for creating _open covers_, which are collections of overlapping sets |
| 10 | + that cover the data space. |
| 11 | +- `learn`: Includes classes compatible with scikit-learn's estimator API based on Mapper. These |
| 12 | + classes can be used in scikit-learn pipelines. |
| 13 | +- `utils`: Provides utility functions for creating spatial indexes. |
| 14 | +- `plot`: Contains functions for visualizing the Mapper graph. |
| 15 | +
|
| 16 | +To use the TDA Mapper algorithm, you can follow these steps: |
| 17 | +
|
| 18 | +Examples |
| 19 | +-------- |
| 20 | +>>> from sklearn.datasets import make_circles |
| 21 | +>>> |
| 22 | +>>> import numpy as np |
| 23 | +>>> from sklearn.decomposition import PCA |
| 24 | +>>> from sklearn.cluster import DBSCAN |
| 25 | +>>> |
| 26 | +>>> from tdamapper.learn import MapperAlgorithm |
| 27 | +>>> from tdamapper.cover import CubicalCover |
| 28 | +>>> from tdamapper.plot import MapperPlot |
| 29 | +>>> |
| 30 | +>>> X, labels = make_circles(n_samples=5000, noise=0.05, factor=0.3, random_state=42) |
| 31 | +>>> y = PCA(2, random_state=42).fit_transform(X) |
| 32 | +>>> |
| 33 | +>>> cover = CubicalCover(n_intervals=10, overlap_frac=0.3) |
| 34 | +>>> clust = DBSCAN() |
| 35 | +>>> graph = MapperAlgorithm(cover, clust).fit_transform(X, y) |
| 36 | +>>> |
| 37 | +>>> fig = MapperPlot(graph, dim=2, seed=42, iterations=60).plot_plotly(colors=labels) |
| 38 | +>>> fig.show(config={"scrollZoom": True}) |
| 39 | +""" |
0 commit comments