A grid of scatter plots showing all pairwise relationships between multiple variables, with histograms or kernel density estimates on the diagonal. This comprehensive visualization enables simultaneous exploration of correlations and distributions across an entire dataset, making it invaluable for understanding multivariate data structure at a glance. Also known as a pairplot or SPLOM (Scatter Plot Matrix).
""" anyplot.ai
scatter-matrix: Scatter Plot Matrix
Library: letsplot 4.9.0 | Python 3.13.13
Quality: 92/100 | Updated: 2026-05-09
"""
import os
import numpy as np
import pandas as pd
from lets_plot import *
from lets_plot.export import ggsave
LetsPlot.setup_html()
THEME = os.getenv("ANYPLOT_THEME", "light")
PAGE_BG = "#FAF8F1" if THEME == "light" else "#1A1A17"
ELEVATED_BG = "#FFFDF6" if THEME == "light" else "#242420"
INK = "#1A1A17" if THEME == "light" else "#F0EFE8"
INK_SOFT = "#4A4A44" if THEME == "light" else "#B8B7B0"
IMPRINT = ["#009E73", "#C475FD", "#4467A3"]
# Data - Iris-like dataset with 4 variables
np.random.seed(42)
n = 150
species = np.repeat(["Setosa", "Versicolor", "Virginica"], n // 3)
sepal_length_setosa = np.random.normal(5.0, 0.35, n // 3)
sepal_width_setosa = np.random.normal(3.4, 0.38, n // 3)
petal_length_setosa = np.random.normal(1.5, 0.17, n // 3)
petal_width_setosa = np.random.normal(0.25, 0.1, n // 3)
sepal_length_versicolor = np.random.normal(5.9, 0.52, n // 3)
sepal_width_versicolor = np.random.normal(2.8, 0.31, n // 3)
petal_length_versicolor = np.random.normal(4.3, 0.47, n // 3)
petal_width_versicolor = np.random.normal(1.3, 0.2, n // 3)
sepal_length_virginica = np.random.normal(6.6, 0.64, n // 3)
sepal_width_virginica = np.random.normal(3.0, 0.32, n // 3)
petal_length_virginica = np.random.normal(5.5, 0.55, n // 3)
petal_width_virginica = np.random.normal(2.0, 0.27, n // 3)
df = pd.DataFrame(
{
"Sepal Length (cm)": np.concatenate([sepal_length_setosa, sepal_length_versicolor, sepal_length_virginica]),
"Sepal Width (cm)": np.concatenate([sepal_width_setosa, sepal_width_versicolor, sepal_width_virginica]),
"Petal Length (cm)": np.concatenate([petal_length_setosa, petal_length_versicolor, petal_length_virginica]),
"Petal Width (cm)": np.concatenate([petal_width_setosa, petal_width_versicolor, petal_width_virginica]),
"Species": species,
}
)
df_plot = df.rename(
columns={
"Sepal Length (cm)": "Sepal Len",
"Sepal Width (cm)": "Sepal Wid",
"Petal Length (cm)": "Petal Len",
"Petal Width (cm)": "Petal Wid",
}
)
variables = ["Sepal Len", "Sepal Wid", "Petal Len", "Petal Wid"]
n_vars = len(variables)
# Build list of plots and their regions for ggbunch
plots = []
regions = []
# Calculate cell dimensions
margin_top = 0.08
margin_bottom = 0.08
margin_left = 0.02
margin_right = 0.02
available_height = 1.0 - margin_top - margin_bottom
available_width = 1.0 - margin_left - margin_right
cell_width = available_width / n_vars
cell_height = available_height / n_vars
# Create plots for the matrix
for i, var_y in enumerate(variables):
for j, var_x in enumerate(variables):
if i == j:
# Diagonal: histogram
p = (
ggplot(df_plot, aes(x=var_x, fill="Species"))
+ geom_histogram(alpha=0.7, bins=15, position="identity")
+ scale_fill_manual(values=IMPRINT)
+ theme_minimal()
+ theme(
axis_title=element_blank(),
axis_text=element_text(size=14, color=INK_SOFT),
legend_position="none",
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
panel_grid_minor=element_blank(),
panel_grid_major=element_line(color=INK_SOFT, size=0.2),
)
)
else:
# Off-diagonal: scatter plot
p = (
ggplot(df_plot, aes(x=var_x, y=var_y, color="Species"))
+ geom_point(size=3.5, alpha=0.7)
+ scale_color_manual(values=IMPRINT)
+ theme_minimal()
+ theme(
axis_title=element_blank(),
axis_text=element_text(size=14, color=INK_SOFT),
legend_position="none",
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
panel_grid_minor=element_blank(),
panel_grid_major=element_line(color=INK_SOFT, size=0.2),
)
)
# Add variable names on bottom edge (last row)
if i == n_vars - 1:
p = p + labs(x=var_x) + theme(axis_title_x=element_text(size=18, color=INK))
# Add variable names on left edge (first column)
if j == 0:
p = p + labs(y=var_y) + theme(axis_title_y=element_text(size=18, color=INK))
plots.append(p)
x_pos = margin_left + j * cell_width
y_pos = margin_top + i * cell_height
regions.append((x_pos, y_pos, cell_width, cell_height, 0, 0))
# Create title plot
title_plot = (
ggplot()
+ geom_blank()
+ ggtitle("scatter-matrix · letsplot · anyplot.ai")
+ theme_void()
+ theme(
plot_title=element_text(size=32, hjust=0.5, face="bold", color=INK),
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
)
)
plots.append(title_plot)
regions.append((0, 0, 1, margin_top, 0, 0))
# Create legend plot
legend_df = pd.DataFrame({"x": [1, 2, 3], "y": [1, 2, 3], "Species": ["Setosa", "Versicolor", "Virginica"]})
legend_plot = (
ggplot(legend_df, aes(x="x", y="y", color="Species"))
+ geom_point(size=8)
+ scale_color_manual(values=IMPRINT)
+ theme_void()
+ theme(
legend_position="bottom",
legend_title=element_text(size=20, color=INK),
legend_text=element_text(size=18, color=INK_SOFT),
legend_direction="horizontal",
legend_background=element_rect(fill=ELEVATED_BG, color=INK_SOFT),
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
)
+ guides(color=guide_legend(override_aes={"size": 10}))
)
plots.append(legend_plot)
regions.append((0.2, 1.0 - margin_bottom, 0.6, margin_bottom, 0, 0))
# Combine into ggbunch with square format
combined = ggbunch(plots, regions) + ggsize(1200, 1200)
# Save with scale for high resolution (target ~3600x3600)
ggsave(combined, filename=f"plot-{THEME}.png", path=".", scale=3)
ggsave(combined, filename=f"plot-{THEME}.html", path=".")
Part of Scatter Plot Matrix on anyplot.ai.