Skip to content

Commit 2bb0ffd

Browse files
github-actions[bot]github-actions[bot]
authored
feat(letsplot): implement scatter-matrix-interactive (#3614)
## Implementation: `scatter-matrix-interactive` - letsplot Implements the **letsplot** version of `scatter-matrix-interactive`. **File:** `plots/scatter-matrix-interactive/implementations/letsplot.py` **Parent Issue:** #3604 --- :robot: *[impl-generate workflow](https://github.com/MarkusNeusinger/pyplots/actions/runs/20870870323)* --------- Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
1 parent 46fdc1d commit 2bb0ffd

2 files changed

Lines changed: 423 additions & 0 deletions

File tree

‎plots/scatter-matrix-interactive/implementations/letsplot.py‎

Lines changed: 209 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,209 @@
1+
""" pyplots.ai
2+
scatter-matrix-interactive: Interactive Scatter Plot Matrix (SPLOM)
3+
Library: letsplot 4.8.2 | Python 3.13.11
4+
Quality: 78/100 | Created: 2026-01-10
5+
"""
6+
7+
import numpy as np
8+
import pandas as pd
9+
from lets_plot import (
10+
LetsPlot,
11+
aes,
12+
element_blank,
13+
element_text,
14+
geom_histogram,
15+
geom_point,
16+
ggbunch,
17+
ggplot,
18+
ggsave,
19+
ggsize,
20+
ggtitle,
21+
labs,
22+
layer_tooltips,
23+
scale_color_manual,
24+
scale_fill_manual,
25+
theme,
26+
theme_minimal,
27+
)
28+
29+
30+
LetsPlot.setup_html()
31+
32+
# Data - Synthetic iris-like dataset (4 numeric variables, 150 points)
33+
np.random.seed(42)
34+
35+
# Generate measurements for three plant species with realistic correlations
36+
n_per_species = 50
37+
38+
# Species A (small flowers): lower values
39+
species_a = pd.DataFrame(
40+
{
41+
"Sepal Length (cm)": np.random.normal(5.0, 0.35, n_per_species),
42+
"Sepal Width (cm)": np.random.normal(3.4, 0.38, n_per_species),
43+
"Petal Length (cm)": np.random.normal(1.5, 0.17, n_per_species),
44+
"Petal Width (cm)": np.random.normal(0.25, 0.10, n_per_species),
45+
"Species": "Setosa",
46+
}
47+
)
48+
49+
# Species B (medium flowers): intermediate values with correlation
50+
base_b = np.random.normal(0, 1, n_per_species)
51+
species_b = pd.DataFrame(
52+
{
53+
"Sepal Length (cm)": 5.9 + 0.5 * base_b + np.random.normal(0, 0.2, n_per_species),
54+
"Sepal Width (cm)": 2.8 + 0.3 * base_b + np.random.normal(0, 0.2, n_per_species),
55+
"Petal Length (cm)": 4.3 + 0.5 * base_b + np.random.normal(0, 0.3, n_per_species),
56+
"Petal Width (cm)": 1.3 + 0.2 * base_b + np.random.normal(0, 0.15, n_per_species),
57+
"Species": "Versicolor",
58+
}
59+
)
60+
61+
# Species C (large flowers): higher values with strong correlation
62+
base_c = np.random.normal(0, 1, n_per_species)
63+
species_c = pd.DataFrame(
64+
{
65+
"Sepal Length (cm)": 6.6 + 0.6 * base_c + np.random.normal(0, 0.3, n_per_species),
66+
"Sepal Width (cm)": 3.0 + 0.3 * base_c + np.random.normal(0, 0.25, n_per_species),
67+
"Petal Length (cm)": 5.6 + 0.6 * base_c + np.random.normal(0, 0.3, n_per_species),
68+
"Petal Width (cm)": 2.0 + 0.3 * base_c + np.random.normal(0, 0.2, n_per_species),
69+
"Species": "Virginica",
70+
}
71+
)
72+
73+
df = pd.concat([species_a, species_b, species_c], ignore_index=True)
74+
75+
# Variables for the scatter matrix (use shorter labels for axis to prevent truncation)
76+
variables = ["Sepal Length (cm)", "Sepal Width (cm)", "Petal Length (cm)", "Petal Width (cm)"]
77+
short_labels = ["Sepal Len.", "Sepal Wid.", "Petal Len.", "Petal Wid."]
78+
n = len(variables)
79+
80+
# Color palette - accessible colors with good contrast on white background
81+
colors = ["#306998", "#E67E22", "#16A085"]
82+
83+
# Create individual plots for the 4x4 matrix
84+
plots = []
85+
for i, var_y in enumerate(variables):
86+
for j, var_x in enumerate(variables):
87+
# Only show axis labels on edges (bottom row for x, left column for y)
88+
show_x_label = i == n - 1
89+
show_y_label = j == 0
90+
91+
# Get short labels for axes
92+
x_label = short_labels[j]
93+
y_label = short_labels[i]
94+
95+
if i == j:
96+
# Diagonal: histogram showing distribution
97+
p = (
98+
ggplot(df, aes(x=var_x, fill="Species"))
99+
+ geom_histogram(alpha=0.7, bins=15, position="identity")
100+
+ scale_fill_manual(values=colors)
101+
+ labs(x=x_label if show_x_label else "", y="")
102+
+ theme_minimal()
103+
+ theme(
104+
axis_title_x=element_text(size=16) if show_x_label else element_blank(),
105+
axis_title_y=element_blank(),
106+
axis_text=element_text(size=13),
107+
legend_position="none",
108+
plot_margin=[5, 5, 5, 5],
109+
)
110+
)
111+
else:
112+
# Off-diagonal: scatter plot with tooltips for interactivity
113+
p = (
114+
ggplot(df, aes(x=var_x, y=var_y, color="Species", fill="Species"))
115+
+ geom_point(
116+
size=4,
117+
alpha=0.7,
118+
shape=21,
119+
tooltips=layer_tooltips()
120+
.line("Species: @Species")
121+
.line(f"{var_x}: @{{{var_x}}}")
122+
.line(f"{var_y}: @{{{var_y}}}"),
123+
)
124+
+ scale_color_manual(values=colors)
125+
+ scale_fill_manual(values=colors)
126+
+ labs(x=x_label if show_x_label else "", y=y_label if show_y_label else "")
127+
+ theme_minimal()
128+
+ theme(
129+
axis_title_x=element_text(size=16) if show_x_label else element_blank(),
130+
axis_title_y=element_text(size=16) if show_y_label else element_blank(),
131+
axis_text=element_text(size=13),
132+
legend_position="none",
133+
plot_margin=[5, 5, 5, 5],
134+
)
135+
)
136+
plots.append(p)
137+
138+
# Calculate regions for ggbunch (4x4 grid)
139+
# Leave space for title at top and legend at bottom
140+
title_height = 0.06
141+
legend_height = 0.06
142+
grid_height = 1.0 - title_height - legend_height
143+
cell_size = grid_height / n
144+
145+
# Title plot - needs a geom layer for lets-plot (use middle dot character)
146+
title_df = pd.DataFrame({"x": [0], "y": [0]})
147+
title_plot = (
148+
ggplot(title_df, aes(x="x", y="y"))
149+
+ geom_point(alpha=0) # Invisible point to satisfy lets-plot layer requirement
150+
+ ggtitle("scatter-matrix-interactive \u00b7 letsplot \u00b7 pyplots.ai")
151+
+ theme_minimal()
152+
+ theme(
153+
plot_title=element_text(size=32, hjust=0.5),
154+
axis_line=element_blank(),
155+
axis_text=element_blank(),
156+
axis_ticks=element_blank(),
157+
axis_title=element_blank(),
158+
panel_grid=element_blank(),
159+
)
160+
)
161+
162+
# Legend plot - separate plot for better control over legend text
163+
legend_df = pd.DataFrame({"x": [1, 2, 3], "y": [0, 0, 0], "Species": ["Setosa", "Versicolor", "Virginica"]})
164+
legend_plot = (
165+
ggplot(legend_df, aes(x="x", y="y", color="Species", fill="Species"))
166+
+ geom_point(size=6, shape=21, alpha=0.8)
167+
+ scale_color_manual(values=colors)
168+
+ scale_fill_manual(values=colors)
169+
+ theme_minimal()
170+
+ theme(
171+
legend_position="bottom",
172+
legend_direction="horizontal",
173+
legend_title=element_text(size=18),
174+
legend_text=element_text(size=16),
175+
axis_line=element_blank(),
176+
axis_text=element_blank(),
177+
axis_ticks=element_blank(),
178+
axis_title=element_blank(),
179+
panel_grid=element_blank(),
180+
)
181+
)
182+
183+
# Build final layout: title + matrix + legend
184+
final_plots = [title_plot]
185+
final_plots.extend(plots)
186+
final_plots.append(legend_plot)
187+
188+
# Define regions for ggbunch
189+
final_regions = []
190+
191+
# Title region (top)
192+
final_regions.append((0, 0, 1, title_height, 0, 0))
193+
194+
# Matrix regions (4x4 grid) - slightly offset from left edge
195+
for idx in range(n * n):
196+
row = idx // n
197+
col = idx % n
198+
x = col * cell_size + 0.02
199+
y = title_height + row * cell_size
200+
final_regions.append((x, y, cell_size, cell_size, 0, 0))
201+
202+
# Legend region (bottom center)
203+
final_regions.append((0.25, 1.0 - legend_height, 0.5, legend_height, 0, 0))
204+
205+
# Combine all plots using ggbunch with square aspect ratio
206+
final_plot = ggbunch(final_plots, final_regions) + ggsize(1200, 1200)
207+
208+
# Save output to current directory (PNG only)
209+
ggsave(final_plot, "plot.png", path=".", scale=3)

0 commit comments

Comments
 (0)