Programming Language Benchmark Raincloud

import numpy as np
import matplotlib.pyplot as plt
import scipy.stats as stats

np.random.seed(42)

# Execution time in milliseconds for a benchmark task
python_times = np.random.lognormal(4.5, 0.4, 80)
java_times = np.random.lognormal(3.2, 0.35, 85)
rust_times = np.random.lognormal(2.0, 0.3, 75)

python_times = np.clip(python_times, 30, 400)
java_times = np.clip(java_times, 10, 100)
rust_times = np.clip(rust_times, 3, 30)

# ANOVA
F_stat, p_value = stats.f_oneway(python_times, java_times, rust_times)

# Colors
BG_COLOR = "#0a0a0f"
GREY_LIGHT = "#444444"
COLOR_SCALE = ["#F5276C", "#27D3F5", "#6CF527"]

fig, ax = plt.subplots(figsize=(10, 6), facecolor=BG_COLOR)
ax.set_facecolor(BG_COLOR)

y_data = [python_times, java_times, rust_times]
positions = [0, 1, 2]
labels = ["Python", "Java", "Rust"]

# Horizontal reference lines
for h in [50, 100, 200, 300]:
    ax.axhline(h, color=GREY_LIGHT, ls=(0, (5, 5)), alpha=0.5, zorder=0)

# Violins
violins = ax.violinplot(y_data, positions=positions, widths=0.45, 
                         bw_method="silverman", showmeans=False, 
                         showmedians=False, showextrema=False)
for pc in violins["bodies"]:
    pc.set_facecolor("none")
    pc.set_edgecolor("white")
    pc.set_linewidth(1.8)
    pc.set_alpha(0.9)

# Boxplots
bp = ax.boxplot(y_data, positions=positions, showfliers=False, showcaps=False,
                medianprops=dict(linewidth=3, color='#F5D327'),
                whiskerprops=dict(linewidth=2, color='#888888'),
                boxprops=dict(linewidth=2, color='#888888'))

# Jittered points
for i, (y, color) in enumerate(zip(y_data, COLOR_SCALE)):
    x_jitter = np.array([i] * len(y)) + stats.t(df=6, scale=0.04).rvs(len(y))
    ax.scatter(x_jitter, y, s=60, color=color, alpha=0.5, zorder=2)

# Mean annotations
means = [y.mean() for y in y_data]
for i, (mean, color) in enumerate(zip(means, COLOR_SCALE)):
    ax.scatter(i, mean, s=200, color='#C82909', zorder=5, edgecolors='white', linewidths=2)
    ax.plot([i, i + 0.3], [mean, mean], ls="dashdot", color="white", zorder=3, lw=1.5)
    ax.text(i + 0.32, mean, f"μ = {mean:.1f}ms", fontsize=11, va="center", color='white',
            bbox=dict(facecolor=BG_COLOR, edgecolor=color, boxstyle="round,pad=0.2", lw=2), zorder=10)

# Comparison brackets
tick_len = 8
# Python vs Java
ax.plot([0, 0, 1, 1], [380-tick_len, 380, 380, 380-tick_len], c="white", lw=1.5)
ax.text(0.5, 385, "p < 0.001 ***", fontsize=10, va="bottom", ha="center", color='#F5D327')
# Python vs Rust
ax.plot([0, 0, 2, 2], [410-tick_len, 410, 410, 410-tick_len], c="white", lw=1.5)
ax.text(1, 415, "p < 0.001 ***", fontsize=10, va="bottom", ha="center", color='#F5D327')

# Styling
ax.spines["right"].set_visible(False)
ax.spines["top"].set_visible(False)
ax.spines["left"].set_color(GREY_LIGHT)
ax.spines["bottom"].set_color(GREY_LIGHT)
ax.tick_params(colors='#888888', length=0)

xlabels = [f"{l}\n(n={len(y_data[i])})" for i, l in enumerate(labels)]
ax.set_xticks(positions)
ax.set_xticklabels(xlabels, size=12, color='white')
ax.set_ylabel("Execution Time (ms)", size=14, color='white', fontweight='bold')

# Title with stats
ax.set_title("Programming Language Performance Benchmark", fontsize=14, color="white", fontweight="bold", pad=20)
ax.text(0.5, -0.12, "F-Welch={F_stat:.1f}, p={p_value:.2e}, ω²=0.89, n=240", transform=ax.transAxes, fontsize=10, color="#6b7280", ha="center")

ax.set_ylim(0, 450)
plt.tight_layout()
plt.show()