Export High-Quality Figures from Python

This guide shows you how to export publication-quality figures from Python using Plotly, ensuring your plots are crisp and professional for papers, presentations, and posters.

Overview

Default Plotly exports are often too low-resolution for publications. This guide explains how to configure high-quality exports for:

• PNG format - Raster images for presentations and web
• SVG format - Vector graphics for publications and posters

The Problem

When you click the camera icon in a Plotly figure, the default export:

• Uses low resolution (often 72-96 DPI)
• May appear pixelated when printed
• Doesn't meet journal requirements (typically 300+ DPI)

The Solution

Configure Plotly's export settings before displaying your figure.

Quick Start

Here's the complete code to enable high-quality exports:

# Define the Plotly configuration
PLOTLY_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',  # Export format: 'png' or 'svg'
        'scale': 10,      # Scale factor (multiplies default resolution)
    }
}

# Show your figure with the configuration
fig.show(config=PLOTLY_CONFIG)

After running this, the camera icon in the Plotly figure will export high-quality images!

Step-by-Step Instructions

Step 1: Import Required Libraries

Make sure you have Plotly installed:

import plotly.graph_objects as go
import plotly.express as px

Step 2: Create Your Figure

Create your plot as usual:

# Example figure
fig = go.Figure(data=go.Scatter(
    x=[1, 2, 3, 4],
    y=[10, 11, 12, 13]
))

fig.update_layout(
    title="My High-Quality Plot",
    xaxis_title="X Axis",
    yaxis_title="Y Axis"
)

Step 3: Define Export Configuration

For PNG (Raster) Export:

PLOTLY_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',    # PNG format
        'scale': 10,        # 10x resolution (very high quality)
    }
}

For SVG (Vector) Export:

PLOTLY_CONFIG = {
    'toImageButtonOptions': {
        'format': 'svg',    # SVG format (scalable vector)
        'scale': 1,         # Scale factor (less important for vectors)
    }
}

Step 4: Display with Configuration

Instead of:

fig.show()  # Default low quality

Use:

fig.show(config=PLOTLY_CONFIG)  # High quality!

Step 5: Export theimage

1. The Plotly figure appears in your Jupyter notebook
2. Hover over the figure - controls appear in the top-right corner
3. Click the camera icon (📷)
4. The high-quality image downloads to your browser's download folder

Configuration Options

Format Selection

PNG - Raster format:

• ✅ Good for presentations and posters
• ✅ Smaller file sizes with high scale factors
• ✅ Widely compatible
• ❌ Can pixelate when zoomed
• ❌ Fixed resolution

SVG - Vector format:

• ✅ Perfect for publications
• ✅ Scales infinitely without quality loss
• ✅ Editable in Illustrator/Inkscape
• ❌ Larger file sizes for complex plots
• ❌ May have font rendering issues in some viewers

Scale Factors

The scale parameter multiplies the default resolution:

Scale	Output Quality	Use Case
1	~96 DPI	Screen viewing only
2	~192 DPI	Web graphics
5	~480 DPI	Presentations
10	~960 DPI	Publications, posters

Recommended: scale=10

For publication quality, use scale=10 with PNG format. This ensures crisp prints even on large posters.

Additional Options

You can add more configuration options:

PLOTLY_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',
        'scale': 10,
        'width': 1920,    # Set explicit width (pixels)
        'height': 1080,   # Set explicit height (pixels)
        'filename': 'my_figure',  # Default filename
    },
    'displayModeBar': True,      # Always show toolbar
    'displaylogo': False,        # Hide Plotly logo
}

Creating a Reusable Configuration

For consistency across all your figures, define configurations once:

# At the top of your notebook
PNG_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',
        'scale': 10,
    }
}

SVG_CONFIG = {
    'toImageButtonOptions': {
        'format': 'svg',
        'scale': 1,
    }
}

# Use throughout your notebook
fig1.show(config=PNG_CONFIG)
fig2.show(config=SVG_CONFIG)
fig3.show(config=PNG_CONFIG)

Best Practices

For Publications

# Vector graphics, editable
PUBLICATION_CONFIG = {
    'toImageButtonOptions': {
        'format': 'svg',
        'scale': 1,
    },
    'displaylogo': False,
}

fig.show(config=PUBLICATION_CONFIG)

Then:

1. Export as SVG
2. Open in vector graphics editor (Illustrator, Inkscape)
3. Adjust fonts, colors, labels as needed
4. Export final version per journal requirements

For Presentations

# High-resolution raster
PRESENTATION_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',
        'scale': 5,  # Balance quality vs file size
        'width': 1920,
        'height': 1080,
    }
}

fig.show(config=PRESENTATION_CONFIG)

For Posters

# Very high resolution for large prints
POSTER_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',
        'scale': 10,  # Maximum quality
    }
}

fig.show(config=POSTER_CONFIG)

Troubleshooting

Export still looks low quality

Problem: Image appears pixelated despite configuration

Solutions:

• Verify you used fig.show(config=PLOTLY_CONFIG) not just fig.show()
• Check the scale factor (scale=10 for high quality)
• Try increasing the scale factor further
• For vectors, use SVG instead of PNG

Export is too large

Problem: File size is very large

Solutions:

• For PNG: Reduce scale factor (try scale=5)
• For SVG: Simplify your plot (fewer data points)
• Consider PNG for complex plots with many points
• Use subsampling for scatter plots with thousands of points

Camera icon doesn't appear

Problem: Can't find the export button

Solutions:

• Hover over the plot to reveal toolbar
• Add 'displayModeBar': True to configuration
• Try 'modeBarButtonsToAdd': ['toImage'] in config
• Verify Plotly is properly installed and up to date

SVG renders incorrectly

Problem: SVG file has missing elements or wrong fonts

Solutions:

• Try PNG export instead
• Update Plotly to latest version
• Simplify plot (remove problematic elements)
• Export PNG at high scale, then vectorize externally

Export filename is generic

Problem: All exports named "newplot.png"

Solutions:

Add filename to configuration:

PLOTLY_CONFIG = {
    'toImageButtonOptions': {
        'format': 'png',
        'scale': 10,
        'filename': 'figure_1_composition_map',
    }
}

Alternative: Programmatic Export

For batch processing or automation, export programmatically:

Install Required Package

pip install kaleido

Export from Code

# PNG export
fig.write_image("figure.png", scale=10)

# SVG export
fig.write_image("figure.svg")

# With size specification
fig.write_image("figure.png", width=1920, height=1080, scale=10)

This bypasses the GUI and generates files directly.

Related Resources

• Plot Combinatorial EDX Data - Creating the figures to export
• Jupyter Analysis Reference - Advanced visualization techniques
• Plotly Documentation - Official Plotly guides

Need Help?

If you encounter export issues:

• Check Plotly version: pip show plotly
• Update Plotly: pip install --upgrade plotly
• Try programmatic export as alternative
• Ask colleagues for working configuration examples
• Consult Plotly community forums