Examples and Tutorials

This section provides comprehensive examples of how to use Demoviz for different types of demographic visualizations.

Baby Growth Chart

Create a simple baby growth chart using baby icons:

import numpy as np
import matplotlib.pyplot as plt
from demoviz import DemoScatter

# Create sample baby health data
ages = [0, 6, 12, 18, 24]  # Age in months
weights = [3.5, 7.0, 9.5, 11.0, 12.5]  # Weight in kg

# Create the plot
fig, ax = plt.subplots(figsize=(8, 6))

# Initialize DemoScatter with baby icons
demo = DemoScatter(icon_size=40, zoom=0.4)

# Plot babies with baby icons
demo.plot(ax, ages, weights, icon_type='baby', colors='lightblue')

# Customize the plot
ax.set_xlabel('Age (months)')
ax.set_ylabel('Weight (kg)')
ax.set_title('Baby Growth Chart')
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.show()

Multi-Category Visualization

Visualize different demographics using different icon types and colors:

import numpy as np
import matplotlib.pyplot as plt
from demoviz import DemoScatter

# Create mixed demographic data
np.random.seed(42)

# Adults (male/female)
adult_ages = np.random.uniform(25, 65, 15)
adult_heights = np.random.normal(170, 10, 15)
adult_genders = np.random.choice(['male', 'female'], 15)

# Babies
baby_ages = np.random.uniform(0, 2, 8)
baby_heights = np.random.normal(75, 10, 8)

# Combine data
all_ages = np.concatenate([adult_ages, baby_ages])
all_heights = np.concatenate([adult_heights, baby_heights])
all_icons = np.concatenate([adult_genders, ['baby'] * 8])

# Create the plot
fig, ax = plt.subplots(figsize=(12, 8))

demo = DemoScatter(icon_size=35, zoom=0.4)

# Plot with different icons and colors
colors = ['red' if icon == 'female' else 'blue' if icon == 'male' else 'lightgreen'
          for icon in all_icons]

demo.plot(ax, all_ages, all_heights, icon_type=all_icons, colors=colors)

# Customize the plot
ax.set_xlabel('Age (years)')
ax.set_ylabel('Height (cm)')
ax.set_title('Multi-Demographics Height Distribution')
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.show()

Model Performance Visualization

Use baby icons to represent model performance in medical AI:

import pandas as pd
import matplotlib.pyplot as plt
from demoviz import DemoScatter
from matplotlib.colors import to_rgba

# Sample model performance data
models = ['Model A', 'Model B', 'Model C', 'Model D', 'Model E']
accuracy = [0.85, 0.78, 0.92, 0.88, 0.95]
auroc = [0.89, 0.81, 0.94, 0.91, 0.97]
model_types = ['RGB', 'XRV', 'RGB', 'XRV', 'XRV']

# Create plot
fig, ax = plt.subplots(figsize=(10, 8))

demo = DemoScatter(icon_size=60, zoom=0.5)

# Color by model type
colors = []
for model_type in model_types:
    if model_type == 'RGB':
        colors.append(to_rgba('#4A90E2'))  # Blue
    else:
        colors.append(to_rgba('#E85D75'))  # Pink

# Plot with baby icons
demo.plot(ax, accuracy, auroc, icon_type='baby', colors=colors)

# Add labels for each model
for i, model in enumerate(models):
    ax.annotate(model, (accuracy[i], auroc[i]),
                xytext=(10, 10), textcoords='offset points',
                fontsize=10, weight='bold')

ax.set_xlabel('Accuracy')
ax.set_ylabel('AUROC')
ax.set_title('Baby Health Prediction Models Performance')
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.show()

Customization Options

Icon Size and Zoom

Control the appearance of icons:

# Small icons
demo_small = DemoScatter(icon_size=20, zoom=0.3)

# Medium icons (default)
demo_medium = DemoScatter(icon_size=40, zoom=0.4)

# Large icons
demo_large = DemoScatter(icon_size=80, zoom=0.6)

Color Schemes

Different ways to color your icons:

# Single color
demo.plot(ax, x, y, icon_type='baby', colors='lightblue')

# Multiple colors
colors = ['red', 'blue', 'green', 'orange', 'purple']
demo.plot(ax, x, y, icon_type='baby', colors=colors)

# RGB tuples
colors = [(0.2, 0.8, 0.2), (0.8, 0.2, 0.2), (0.2, 0.2, 0.8)]
demo.plot(ax, x, y, icon_type='baby', colors=colors)

# Matplotlib color conversion
from matplotlib.colors import to_rgba
colors = [to_rgba('#4A90E2'), to_rgba('#E85D75')]
demo.plot(ax, x, y, icon_type='baby', colors=colors)

Icon Types

Available icon types and their aliases:

# Baby icons
demo.plot(ax, x, y, icon_type='baby')
demo.plot(ax, x, y, icon_type='child')
demo.plot(ax, x, y, icon_type='infant')

# Male icons
demo.plot(ax, x, y, icon_type='male')
demo.plot(ax, x, y, icon_type='person')

# Female icons
demo.plot(ax, x, y, icon_type='female')
demo.plot(ax, x, y, icon_type='woman')

Performance Tips

  1. Cache Management: The library automatically caches rendered icons. Use demo.clear_cache() if you need to free memory.

  2. Icon Sizes: Larger icons (>100px) may impact performance. Consider using appropriate zoom levels instead.

  3. Color Formats: RGB tuples are processed faster than hex strings or named colors.

  4. Batch Processing: When creating multiple plots, reuse the same DemoScatter instance to benefit from caching.

Troubleshooting

Common Issues

Icons appear too small:

  • Increase the zoom parameter

  • Increase the icon_size parameter

Icons don’t appear:

  • Check that the icon type is supported

  • Verify that axis limits include your data points

  • Ensure colors are in the correct format

Performance issues:

  • Use smaller icon sizes

  • Clear the cache periodically with demo.clear_cache()

  • Consider using fewer data points for initial testing