Getting started with Mapper

In this notebook we explore a few of the core features included in giotto-tda's implementation of the Mapper algorithm.

If you are looking at a static version of this notebook and would like to run its contents, head over to github.

Useful references

License: AGPLv3

Import libraries

In [ ]:
# Data wrangling
import numpy as np
import pandas as pd  # Not a requirement of giotto-tda, but is compatible with the gtda.mapper module

# Data viz
from gtda.plotting import plot_point_cloud

# TDA magic
from gtda.mapper import (
from gtda.mapper.utils.visualization import set_node_sizeref

# ML tools
from sklearn import datasets
from sklearn.cluster import DBSCAN
from sklearn.decomposition import PCA

Generate and visualise data

As a simple example, let's generate a two-dimensional point cloud of two concentric circles. The goal will be to examine how Mapper can be used to generate a topological graph that captures the salient features of the data.

In [ ]:
data, _ = datasets.make_circles(n_samples=5000, noise=0.05, factor=0.3, random_state=42)


Configure the Mapper pipeline

Given a dataset ${\cal D}$ of points $x \in \mathbb{R}^n$, the basic steps behind Mapper are as follows:

  1. Map ${\cal D}$ to a lower-dimensional space using a filter function $f: \mathbb{R}^n \to \mathbb{R}^m$. Common choices for the filter function include projection onto one or more axes via PCA or density-based methods. In giotto-tda, you can import a variety of filter functions as follows:
from gtda.mapper.filter import FilterFunctionName
  1. Construct a cover of the filter values ${\cal U} = (U_i)_{i\in I}$, typically in the form of a set of overlapping intervals which have constant length. As with the filter, a choice of cover can be imported as follows:
from gtda.mapper.cover import CoverName
  1. For each interval $U_i \in {\cal U}$ cluster the points in the preimage $f^{-1}(U_i)$ into sets $C_{i,1}, \ldots , C_{i,k_i}$. The choice of clustering algorithm can be any of scikit-learn's clustering methods or an implementation of agglomerative clustering in giotto-tda:
# scikit-learn method
from sklearn.cluster import ClusteringAlgorithm
# giotto-tda method
from gtda.mapper.cluster import FirstSimpleGap
  1. Construct the topological graph whose vertices are the cluster sets $(C_{i,j})_{i\in I, j \in \{1,\ldots,k_i\}}$ and an edge exists between two nodes if they share points in common: $C_{i,j} \cap C_{k,l} \neq \emptyset$. This step is handled automatically by giotto-tda.

These four steps are implemented in the MapperPipeline object that mimics the Pipeline class from scikit-learn. We provide a convenience function make_mapper_pipeline() that allows you to pass the choice of filter function, cover, and clustering algorithm as arguments. For example, to project our data onto the $x$- and $y$-axes, we could setup the pipeline as follows:

In [ ]:
# Define filter function - can be any scikit-learn transformer
filter_func = Projection(columns=[0, 1])
# Define cover
cover = CubicalCover(n_intervals=10, overlap_frac=0.3)
# Choose clustering algorithm - default is DBSCAN
clusterer = DBSCAN()

# Configure parallelism of clustering step
n_jobs = 1

# Initialise pipeline
pipe = make_mapper_pipeline(

Visualise the Mapper graph

With the Mapper pipeline at hand, it is now a simple matter to visualise it. To warm up, let's examine the graph in two-dimensions using the default arguments of giotto-tda's plotting function:

In [ ]:
fig = plot_static_mapper_graph(pipe, data){'scrollZoom': True})

From the figure we can see that we have captured the salient topological features of our underlying data, namely two holes!

Configure the coloring of the Mapper graph

By default, the nodes of the Mapper graph are colored by the mean value of the points that belong to a given node. However, in this example it is more instructive to colour by the $x$- and $y$-axes. This can be achieved by toggling the color_by_columns_dropdown, which calculates the coloring for each column in the input data array. At the same time, let's configure the choice of colorscale:

In [ ]:
plotly_kwargs = {"node_trace_marker_colorscale": "Blues"}
fig = plot_static_mapper_graph(
    pipe, data, color_by_columns_dropdown=True, plotly_kwargs=plotly_kwargs
){'scrollZoom': True})

In the dropdown menu, the entry color_variable refers to a user-defined quantity to color by - by default it is the average value of the points in each node. In general, one can configure this quantity to be an array, a scikit-learn transformer, or a list of indices to select from the data. For example, coloring by a PCA component can be implemented as follows:

In [ ]:
# Initialise estimator to color graph by
pca = PCA(n_components=1).fit(data)

fig = plot_static_mapper_graph(
    pipe, data, color_by_columns_dropdown=True, color_variable=pca
){'scrollZoom': True})

Pass a pandas DataFrame as input

It is also possible to feed plot_static_mapper_graph() a pandas DataFrame:

In [ ]:
data_df = pd.DataFrame(data, columns=["x", "y"])

Before plotting we need to update the Mapper pipeline to know about the projection onto the column names. This can be achieved using the set_params() method as follows:

In [ ]:
pipe.set_params(filter_func=Projection(columns=["x", "y"]));
In [ ]:
fig = plot_static_mapper_graph(pipe, data_df, color_by_columns_dropdown=True){'scrollZoom': True})

Change the layout algorithm

By default, plot_static_mapper_graph() uses the Kamada–Kawai algorithm for the layout; however any of the layout algorithms defined in python-igraph are supported (see here for a list of possible layouts). For example, we can switch to the Fruchterman–Reingold layout as follows:

In [ ]:
# Reset back to numpy projection
pipe.set_params(filter_func=Projection(columns=[0, 1]));
In [ ]:
fig = plot_static_mapper_graph(
    pipe, data, layout="fruchterman_reingold", color_by_columns_dropdown=True
){'scrollZoom': True})

Change the layout dimension

It is also possible to visualise the Mapper graph in 3-dimensions by configuring the layout_dim argument:

In [ ]:
fig = plot_static_mapper_graph(pipe, data, layout_dim=3, color_by_columns_dropdown=True){'scrollZoom': True})

Run the Mapper pipeline

Behind the scenes of plot_static_mapper_graph() is a MapperPipeline object pipe that can be used like a typical scikit-learn estimator. For example, to extract the underlying graph data structure we can do the following:

In [ ]:
graph = pipe.fit_transform(data)

The resulting graph is an python-igraph object that contains metadata that is stored in the form of dictionaries. We can access this data as follows:

In [ ]:

Here node_id is a globally unique node identifier used to construct the graph, while pullback_set_label and partial_cluster_label refer to the interval and cluster sets described above. The node_elements refers to the indices of our original data that belong to each node. For example, to find which points belong to the first node of the graph we can access the desired data as follows:

In [ ]:
node_id, node_elements = (
In [ ]:
    "Node Id: {}, \nNode elements: {}, \nData points: {}".format(
        node_id[0], node_elements[0], data[node_elements[0]]

The node_elements are handy for situations when we want to customise e.g. the size of the node scale. In this example, we use the utility function set_node_sizeref() and pass the function as a plotly argument:

In [ ]:
# Configure scale for node sizes
plotly_kwargs = {
    "node_trace_marker_sizeref": set_node_sizeref(node_elements, node_scale=30)
fig = plot_static_mapper_graph(
){'scrollZoom': True})

The resulting graph is much easier to decipher with the enlarged node scaling!

Creating custom filter functions

In some cases, the list of filter functions provided in or scikit-learn may not be sufficient for the task at hand. In such cases, one can pass any callable to the pipeline that acts row-wise on the input data. For example, we can project by taking the sum of the $(x,y)$ coordinates as follows:

In [ ]:
filter_func = np.sum

pipe = make_mapper_pipeline(
In [ ]:
fig = plot_static_mapper_graph(pipe, data, plotly_kwargs=None){'scrollZoom': True})

Visualise the 2D Mapper graph interactively (Live Jupyter session needed)

In general, building useful Mapper graphs requires some iteration through the various parameters in the cover and clustering algorithm. To simplify that process, giotto-tda provides an interactive figure that can be configured in real time.

If invalid parameters are selected, the Show logs checkbox can be used to see what went wrong.

To see the interactive output, please download the notebook from github and execute it locally.

In [ ]:
pipe = make_mapper_pipeline()

# Generate interactive widget
plot_interactive_mapper_graph(pipe, data, color_by_columns_dropdown=True)