📊 Try it now:
Jupyter Notebook to Slides ↗ | All Tools ↗

Convert Jupyter Notebook to Slides

Jupyter Notebooks are the standard for data science workflows — combining code, visualizations, and narrative. But when it's time to present your analysis to stakeholders, a raw .ipynb file won't do. This guide covers every method for turning notebooks into polished slides.

Why Convert Notebooks to Presentations?

Audience	What They Need
Technical team	Code + results + methodology
Management	Key findings + charts + recommendations
Clients	Insights + visuals, no code
Conference	Narrative flow + key figures

A single notebook can produce different presentations depending on which cells you include.

Method 1: Sharayeh AI Converter (Recommended)

The fastest approach — upload your .ipynb and get a professional .pptx in seconds.

How It Works

Go to Jupyter Notebook to Slides
Upload your .ipynb file
Choose what to include:
- ✅ Markdown cells → text slides
- ✅ Code output / charts → figure slides
- ⬜ Raw code cells (optional — toggle on for technical audiences)
Select a template (clean academic, corporate, or dark theme)
Download your .pptx

What the AI Does

Converts markdown headings into slide titles
Places matplotlib / seaborn / plotly charts as high-res images
Formats pandas DataFrames as PowerPoint tables
Creates speaker notes from your markdown explanations
Handles LaTeX equations in markdown cells

Method 2: Jupyter Built-in Slideshow (nbconvert)

Jupyter has a built-in slideshow feature using RISE / nbconvert:

jupyter nbconvert --to slides notebook.ipynb

Pros:

Built into Jupyter
Outputs HTML slides (Reveal.js)

Cons:

No .pptx output — HTML only
Requires manual cell metadata tagging (Slide / Sub-Slide / Fragment)
Limited design options
Not suitable for email or offline sharing

Method 3: Export to PDF, Then Convert

File → Export as → PDF in Jupyter
Upload the PDF to PDF to PowerPoint

Pros: Works with any notebook
Cons: Loses editability — slides contain images of content

Comparison of Methods

Feature	Sharayeh AI	nbconvert	PDF → PPTX
Output format	.pptx	HTML	.pptx
Editable slides	✅	❌	⚠️ Limited
Chart quality	High-res PNG	Browser-rendered	Depends on PDF
Code formatting	Syntax highlighted	As-is	Image
Speaker notes	✅ Auto-generated	❌	❌
Design templates	✅ Multiple	❌ Basic	❌
Setup required	None (web tool)	Terminal command	Two-step

Best Practices for Data Science Presentations

Structure Your Notebook for Slides

Before converting, organize your notebook:

Title cell (H1 markdown) → becomes title slide
Introduction (H2) → context and objective
Data overview (H2) → key stats, .describe() output
Methodology (H2) → approach, model choice
Results (H2) → charts, metrics, tables
Conclusions (H2) → findings and recommendations

Optimize Visualizations

Set figure size to 10×6 inches for slide readability
Use 24pt font for axis labels and titles
Export at 150+ DPI for crisp rendering
Add gridlines for data plots
Use colorblind-friendly palettes (e.g., viridis, cividis)

import matplotlib.pyplot as plt
plt.rcParams.update({
    'figure.figsize': (10, 6),
    'font.size': 14,
    'axes.titlesize': 18,
    'savefig.dpi': 150
})

Handle Large DataFrames

Don't put a 100-row DataFrame on a slide. Instead:

Show .head(5) or .describe()
Highlight the key rows/columns
Use a chart instead of a table when possible

Handling Special Content

Interactive Plotly Charts

Plotly charts become static images in .pptx. For interactive presentations, consider:

Sharing a live notebook link alongside your deck
Using the HTML slideshow option (nbconvert)
Exporting Plotly as high-res PNG: fig.write_image("chart.png", scale=2)

LaTeX Equations

Equations in markdown cells ( $E = mc^2$ ) are rendered as images in the slides. For complex equations, ensure they render correctly in your notebook first.

Code Cells

For technical audiences, showing code can be valuable:

Keep code cells short (< 10 lines per slide)
Remove debugging / exploratory code
Add comments explaining the "why" not the "what"

Common Use Cases

Kaggle Competition Presentation

Share your approach and leaderboard results:

EDA highlights → 3–4 chart slides
Feature engineering → methodology slide
Model comparison → table slide
Final score → results slide

Client Data Analysis Report

Focus on insights, not code:

Toggle off code cells
Lead with business metrics
End with recommendations and next steps

Academic Research Presentation

Include methodology details:

Show key code snippets
Include statistical tests and p-values
Reference data sources

Frequently Asked Questions

Can I convert Google Colab notebooks?

Yes. Download your Colab notebook as .ipynb (File → Download → Download .ipynb), then upload to Sharayeh.

Does it support R Notebooks?

Currently optimized for Python notebooks. For R, export to HTML or PDF first, then convert.

What about notebooks with widgets (ipywidgets)?

Interactive widgets are captured as their static state at the time of export. For dynamic content, consider recording a screen capture.

Can I choose which cells become slides?

Yes. You can tag cells with slide metadata in Jupyter (View → Cell Toolbar → Slideshow), and the converter respects those tags.

Convert Jupyter Notebook to Slides: Complete Guide (2026)