Implementation:DistrictDataLabs Yellowbrick MissingValuesDispersion

Knowledge Sources	DistrictDataLabs_Yellowbrick
Domains	Data_Quality, Visualization
Last Updated	2026-02-08 05:00 GMT

Overview

Concrete tool for visualizing the spatial distribution of missing values across features and samples, provided by the Yellowbrick contrib module.

Description

The MissingValuesDispersion renders a scatter-style plot showing the exact locations of missing (NaN) values in a dataset. Each feature is a row, and markers indicate which sample indices have missing values. This reveals patterns in missingness such as block-missing or random-missing structures.

Usage

Import this visualizer when you need to understand the spatial pattern of missing data, not just the counts. It complements MissingValuesBar by showing where values are missing rather than how many.

Code Reference

Source Location

• Repository: DistrictDataLabs_Yellowbrick
• File: yellowbrick/contrib/missing/dispersion.py
• Lines: 1-235

Signature

class MissingValuesDispersion(MissingDataVisualizer):
    def __init__(self, alpha=0.5, marker="|", classes=None, **kwargs):
        """Missing values dispersion plot visualizer."""

def missing_dispersion(X, y=None, ax=None, classes=None, alpha=0.5, marker="|", **kwargs):
    """Quick method for one-off missing values dispersion visualization."""

Import

from yellowbrick.contrib.missing import MissingValuesDispersion
from yellowbrick.contrib.missing.dispersion import missing_dispersion

I/O Contract

Inputs

Name	Type	Required	Description
X	array-like or DataFrame	Yes	Feature data with potential NaN values
y	array-like	No	Target labels for coloring
alpha	float	No	Marker transparency (default: 0.5)
marker	str	No	")

Outputs

Name	Type	Description
ax	matplotlib.Axes	Axes with dispersion scatter plot

Usage Examples

import numpy as np
import pandas as pd
from yellowbrick.contrib.missing import MissingValuesDispersion

df = pd.DataFrame({
    "A": [1, np.nan, 3, np.nan, 5, 6, np.nan, 8],
    "B": [np.nan, 2, 3, 4, 5, np.nan, 7, 8],
    "C": [1, 2, np.nan, 4, np.nan, 6, 7, np.nan],
})

viz = MissingValuesDispersion()
viz.fit(df.values)
viz.show()

Related Pages

• Environment:DistrictDataLabs_Yellowbrick_Python_Scikit_Learn_Environment