A scatter plot displaying high-dimensional data projected into 2D space using non-linear dimensionality reduction techniques such as t-SNE or UMAP. Points are colored by cluster or class label, revealing groupings and latent structure in the data. This is a standard visualization in machine learning for exploring embeddings, single-cell RNA-seq data, and NLP document clustering, helping practitioners verify that learned representations capture meaningful distinctions.

""" anyplot.ai
scatter-embedding: t-SNE and UMAP Embedding Visualization
Library: plotnine 0.15.7 | Python 3.13.14
Quality: 90/100 | Updated: 2026-08-11
"""
import os
import numpy as np
import pandas as pd
from plotnine import (
aes,
coord_fixed,
element_blank,
element_line,
element_rect,
element_text,
geom_label,
geom_point,
ggplot,
labs,
scale_color_manual,
stat_ellipse,
theme,
)
from sklearn.datasets import make_blobs
from sklearn.manifold import TSNE
# Theme tokens
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 palette (canonical order, first series always brand green)
IMPRINT = ["#009E73", "#C475FD", "#4467A3", "#BD8233", "#AE3030", "#2ABCCD"]
# Data: single-cell RNA-seq-style clustering scenario (bioinformatics application)
np.random.seed(42)
cell_types = ["T cells", "B cells", "Monocytes", "NK cells", "Dendritic cells", "Neutrophils"]
n_clusters = len(cell_types)
X, y = make_blobs(n_samples=1200, centers=n_clusters, cluster_std=1.2, random_state=42)
tsne = TSNE(n_components=2, perplexity=30, random_state=42)
embedding = tsne.fit_transform(X)
df = pd.DataFrame(
{
"tsne_1": embedding[:, 0],
"tsne_2": embedding[:, 1],
"Cluster": pd.Categorical([cell_types[i] for i in y], categories=cell_types),
}
)
centroids = df.groupby("Cluster", observed=True)[["tsne_1", "tsne_2"]].mean().reset_index()
# Plot
anyplot_theme = theme(
figure_size=(8, 4.5),
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
panel_grid_major=element_line(color=INK, size=0.3, alpha=0.10),
panel_grid_minor=element_blank(),
panel_border=element_blank(),
axis_line_x=element_line(color=INK_SOFT, size=0.8),
axis_line_y=element_line(color=INK_SOFT, size=0.8),
axis_title=element_text(color=INK, size=10),
axis_text=element_blank(),
axis_ticks=element_blank(),
plot_title=element_text(color=INK, size=12),
plot_subtitle=element_text(color=INK_SOFT, size=8),
legend_background=element_rect(fill=ELEVATED_BG, color=INK_SOFT),
legend_text=element_text(color=INK_SOFT, size=8),
legend_title=element_text(color=INK, size=9),
legend_key=element_rect(fill=PAGE_BG),
legend_margin=10,
)
plot = (
ggplot(df, aes(x="tsne_1", y="tsne_2", color="Cluster"))
+ geom_point(size=1.6, alpha=0.7)
+ stat_ellipse(level=0.68, type="t", size=0.5, alpha=0.6, linetype="dashed", show_legend=False)
+ geom_label(
data=centroids,
mapping=aes(x="tsne_1", y="tsne_2", label="Cluster"),
inherit_aes=False,
size=3.0,
color=INK,
fill=ELEVATED_BG,
boxcolor=INK_SOFT,
fontweight="bold",
show_legend=False,
)
+ scale_color_manual(values=IMPRINT)
+ coord_fixed(ratio=1)
+ labs(
title="scatter-embedding · python · plotnine · anyplot.ai",
subtitle="t-SNE (perplexity=30) · 1200 cells · 6 clusters",
x="t-SNE Dimension 1",
y="t-SNE Dimension 2",
)
+ anyplot_theme
)
# Save
plot.save(f"plot-{THEME}.png", dpi=400, width=8, height=4.5, units="in")
Part of t-SNE and UMAP Embedding Visualization on anyplot.ai.