Implementation:DistrictDataLabs Yellowbrick ScatterVisualizer
| Knowledge Sources | |
|---|---|
| Domains | Feature_Analysis, Visualization |
| Last Updated | 2026-02-08 05:00 GMT |
Overview
Concrete tool for bivariate scatter plot visualization with class-based coloring, provided by the Yellowbrick contrib module.
Description
The ScatterVisualizer creates a 2D scatter plot of two features from a dataset, coloring data points by their target class labels. It supports DataFrames, structured arrays, and numpy arrays as input. Each class is rendered in a distinct color with optional custom markers and alpha transparency.
Usage
Import this visualizer when you need a quick bivariate scatter plot with class-colored points for exploratory data analysis. It is the contrib alternative to Yellowbrick's JointPlot for simple bivariate feature inspection.
Code Reference
Source Location
- Repository: DistrictDataLabs_Yellowbrick
- File: yellowbrick/contrib/scatter.py
- Lines: 1-359
Signature
class ScatterVisualizer(DataVisualizer):
def __init__(
self,
ax=None,
features=None,
classes=None,
color=None,
colormap=None,
markers=None,
alpha=1.0,
**kwargs,
):
"""Bivariate feature visualization using Cartesian coordinates."""
def scatterviz(
X, y=None, ax=None, features=None, classes=None,
color=None, colormap=None, markers=None, alpha=1.0, **kwargs,
):
"""Quick method for one-off scatter visualization."""
ScatterViz = ScatterVisualizer # Alias
Import
from yellowbrick.contrib.scatter import ScatterVisualizer, ScatterViz, scatterviz
I/O Contract
Inputs
| Name | Type | Required | Description |
|---|---|---|---|
| X | array-like (n, 2+) | Yes | Feature matrix (first 2 columns used) |
| y | array-like | No | Target class labels for coloring |
| features | list of str | No | Feature names for axis labels |
| classes | list of str | No | Human-readable class names |
| markers | list of str | No | Marker styles per class |
| alpha | float | No | Point transparency (default: 1.0) |
Outputs
| Name | Type | Description |
|---|---|---|
| ax | matplotlib.Axes | Axes with scatter plot |
Usage Examples
from sklearn.datasets import load_iris
from yellowbrick.contrib.scatter import ScatterViz
X, y = load_iris(return_X_y=True)
viz = ScatterViz(features=["sepal length", "sepal width"], classes=["setosa", "versicolor", "virginica"])
viz.fit(X[:, :2], y)
viz.show()