Colormaps Styling

Colormaps & Styling Skill

Overview

Master color management and visual styling with Colorcet and theme customization. Select appropriate colormaps, create accessible visualizations, and apply consistent application styling.

What is Colorcet?

Colorcet provides perceptually uniform colormaps designed for scientific visualization:

• Perceptually uniform: Changes in data correspond to proportional visual changes
• Colorblind-friendly: Palettes designed for accessibility
• Purpose-built: Specific colormaps for different data types
• HoloViz integration: Seamless use across HoloViews, Panel, and Bokeh

Quick Start

Installation

pip install colorcet

Basic Usage

import colorcet as cc
from colorcet import cm
import holoviews as hv

hv.extension('bokeh')

# Use a colormap
data.hvplot.scatter('x', 'y', c='value', cmap=cm['cet_goertzel'])

Core Concepts

1. Colormap Categories

Sequential: Single hue, increasing intensity

# Blues, greens, reds, grays
data.hvplot('x', 'y', c='value', cmap=cm['cet_blues'])

Diverging: Two hues from center point

# Emphasize positive/negative
data.hvplot('x', 'y', c='value', cmap=cm['cet_coolwarm'])

Categorical: Distinct colors for categories

# Qualitative data
data.hvplot('x', 'y', c='category', cmap=cc.palette['tab10'])

Cyclic: Wraps around for angular data

# Angles, directions, phases
data.hvplot('x', 'y', c='angle', cmap=cm['cet_cyclic_c1'])

See: Colormap Reference for complete catalog

2. Accessibility

Colorblind-safe palettes:

# Deuteranopia (red-green)
cmap=cm['cet_d4']

# Protanopia (red-green)
cmap=cm['cet_p3']

# Tritanopia (blue-yellow)
cmap=cm['cet_t10']

# Grayscale-safe
cmap=cm['cet_gray_r']

See: Accessibility Guide for comprehensive guidelines

3. Colormap Selection Guide

4. HoloViews Styling

import holoviews as hv

# Apply colormap
scatter = hv.Scatter(data, 'x', 'y', vdims=['value']).opts(
    color=hv.dim('value').norm(),
    cmap=cm['cet_goertzel'],
    colorbar=True,
    width=600,
    height=400
)

# Style options
scatter.opts(
    size=5,
    alpha=0.7,
    tools=['hover'],
    title='My Plot'
)

See: HoloViews Styling for advanced customization

5. Panel Themes

import panel as pn

# Apply theme
pn.extension(design='material')

# Custom theme
pn.config.theme = 'dark'

# Accent color
template = pn.template.FastListTemplate(
    title='My App',
    accent='#00aa41'
)

See: Panel Themes for theme customization

Common Patterns

Pattern 1: Heatmap with Diverging Colormap

import holoviews as hv
from colorcet import cm

heatmap = hv.HeatMap(data, ['x', 'y'], 'value').opts(
    cmap=cm['cet_coolwarm'],
    colorbar=True,
    width=600,
    height=400,
    tools=['hover']
)

Pattern 2: Categorical Color Assignment

import panel as pn
from colorcet import palette

categories = ['A', 'B', 'C', 'D']
colors = palette['tab10'][:len(categories)]

color_map = dict(zip(categories, colors))
plot = data.hvplot('x', 'y', c='category', cmap=color_map)

Pattern 3: Consistent App Styling

import panel as pn

# Set global theme
pn.extension(design='material')

# Custom CSS
pn.config.raw_css.append("""
.card {
    border-radius: 10px;
    box-shadow: 0 2px 4px rgba(0,0,0,0.1);
}
""")

# Accent color throughout
accent = '#00aa41'
template = pn.template.FastListTemplate(
    title='My Dashboard',
    accent=accent
)

Pattern 4: Responsive Colorbar

from holoviews import opts

plot = data.hvplot.scatter('x', 'y', c='value', cmap=cm['cet_blues']).opts(
    colorbar=True,
    colorbar_opts={
        'title': 'Value',
        'width': 10,
        'ticker': {'desired_num_ticks': 5}
    }
)

Pattern 5: Colorblind-Safe Visualization

from colorcet import cm

# Use colorblind-safe diverging palette
plot = data.hvplot('x', 'y', c='value', cmap=cm['cet_d4']).opts(
    title='Colorblind-Safe Visualization',
    width=600,
    height=400
)

# Alternative: Use patterns/hatching
plot.opts(hatch_pattern='/')

Best Practices

1. Match Colormap to Data Type

# ✅ Good: Sequential for positive values
temp_plot = data.hvplot(c='temperature', cmap=cm['cet_fire'])

# ✅ Good: Diverging for centered data
correlation = data.hvplot(c='correlation', cmap=cm['cet_coolwarm'])

# ❌ Bad: Rainbow/jet colormap (not perceptually uniform)
bad_plot = data.hvplot(c='value', cmap='jet')  # Avoid!

2. Consider Accessibility

# ✅ Good: Colorblind-safe
plot = data.hvplot(c='value', cmap=cm['cet_d4'])

# ✅ Good: Add patterns for print/grayscale
plot.opts(hatch_pattern='/')

# ✅ Good: Test in grayscale
plot.opts(cmap=cm['cet_gray_r'])

3. Consistent Styling

# ✅ Good: Define color scheme once
COLORS = {
    'primary': '#00aa41',
    'secondary': '#616161',
    'accent': '#ff6f00'
}

# Use throughout application
pn.template.FastListTemplate(accent=COLORS['primary'])

4. Meaningful Labels

# ✅ Good: Descriptive colorbar
plot.opts(
    colorbar=True,
    colorbar_opts={'title': 'Temperature (°C)'}
)

# ❌ Bad: No context
plot.opts(colorbar=True)

5. Performance with Large Data

# For large datasets, limit colormap resolution
plot.opts(
    cmap=cm['cet_goertzel'],
    color_levels=256  # Reduce if performance issues
)

Configuration

Global Colormap Defaults

import holoviews as hv
from colorcet import cm

# Set default colormap
hv.opts.defaults(
    hv.opts.Image(cmap=cm['cet_goertzel']),
    hv.opts.Scatter(cmap=cm['cet_blues'])
)

Theme Configuration

import panel as pn

# Material design
pn.extension(design='material')

# Dark mode
pn.config.theme = 'dark'

# Custom theme JSON
pn.config.theme_json = {
    'palette': {
        'primary': '#00aa41',
        'secondary': '#616161'
    }
}

Troubleshooting

Colormap Not Showing

# Check if colormap imported
from colorcet import cm
print(cm['cet_goertzel'])  # Should print colormap

# Verify data range
print(data['value'].min(), data['value'].max())

# Explicit normalization
plot.opts(color=hv.dim('value').norm())

Colors Look Wrong

• Issue: Perceptual non-uniformity
• Solution: Use Colorcet instead of matplotlib defaults

# ❌ Avoid
cmap='jet', cmap='rainbow'

# ✅ Use
cmap=cm['cet_goertzel'], cmap=cm['cet_fire']

Theme Not Applying

# Ensure extension loaded with design
pn.extension(design='material')

# Check theme setting
print(pn.config.theme)  # 'default' or 'dark'

# Reload page after theme change

Progressive Learning Path

Level 1: Basics

1. Install Colorcet
2. Use basic colormaps
3. Apply to plots

Resources:

• Quick Start (this doc)
• Colormap Reference

Level 2: Accessibility

1. Understand colormap categories
2. Choose appropriate maps
3. Test for colorblindness

Resources:

• Accessibility Guide

Level 3: Advanced Styling

1. Customize HoloViews opts
2. Create custom themes
3. Consistent branding

Resources:

• HoloViews Styling
• Panel Themes

Additional Resources

Documentation

• Colormap Reference - Complete colormap catalog
• Accessibility Guide - Colorblind-friendly design
• HoloViews Styling - Advanced customization
• Panel Themes - Theme and branding

External Links

• Colorcet Documentation
• Colorcet Gallery
• Color Universal Design
• WCAG Color Contrast

Use Cases

Scientific Visualization

• Temperature maps
• Density plots
• Correlation matrices
• Geospatial data

Data Dashboards

• KPI indicators
• Time series
• Category comparison
• Status displays

Accessibility

• Colorblind-friendly visualizations
• Print-safe graphics
• High-contrast displays
• Grayscale compatibility

Branding

• Corporate colors
• Consistent styling
• Custom themes
• Professional appearance

Summary

Colorcet provides perceptually uniform, accessible colormaps for scientific visualization.

Key principles:

• Match colormap to data type
• Choose colorblind-safe palettes
• Use perceptually uniform maps
• Maintain consistent styling
• Test accessibility

Ideal for:

• Scientific visualizations
• Accessible dashboards
• Professional applications
• Print publications

Colormap selection:

• Sequential: Single channel data
• Diverging: Centered data (±)
• Categorical: Qualitative categories
• Cyclic: Angular/periodic data

Related Skills

• Data Visualization - HoloViews visualization patterns
• Panel Dashboards - Dashboard styling and themes
• Plotting Fundamentals - Basic plotting with hvPlot