Skip to content
Get packs

Guide · Open source

How to use ML Visuals: reuse ML figures in slides and papers

Updated · written with AI from the project's own documentation and checked against our catalogue data

ML Visuals is a free, community-made collection of over 100 machine learning figures and templates, maintained as Google Slides. This guide explains where to find the visuals, how to copy, customize and download them, and how to credit their authors. By the end you will know how to reuse a figure in a presentation, blog post or paper, and how to add one of your own.

ML Visualsdair-ai/ml-visuals · ★ 17.4k · MIT

Quick start

  1. 1Open the slides. Open the ML Visuals presentation using the Version 1.0 link in the README.
  2. 2Find your figure. Browse the slides for the figure that matches your topic.
  3. 3Copy or customize. Copy the figure into your work, or make a full copy of the deck to edit.
  4. 4Download the file. Go to File, then Download, and choose the format you need.
  5. 5Credit the author. Look up the author in the slide notes and cite them where you use the figure.
  6. 6Request edits or new figures. Ask for edit access or open a GitHub issue if you need a change or a new figure.

What is ML Visuals and what does it contain?

ML Visuals is a collaborative effort from the dair.ai community to improve science communication in machine learning by providing free, professional visuals and figures. The source describes it as containing over 100 figures, all contributed by the open community.

The visuals are kept in Google Slides, so you open them in a browser rather than installing a program. The project's own description calls them free and open artifacts and templates that you can download, copy, distribute, reuse and customize.

Open the slides and copy a figure

Open the ML Visuals presentation from the link in the project README. The README points to Version 1.0 of the slides as the current version. Find the figure you need and copy it into your own slides or document, or make your own copy of the whole deck to edit freely.

If you only want to view the slides and need to change them, use the request edit access option under the view-only toolbar in Google Slides. The README also lists an email address for edit permission requests, which you can use instead.

  • Use the link to Version 1.0 from the README to open the slides.
  • Choose File, then Make a copy to customize the whole deck.
  • Use the request edit access option if you need to change the shared slides.

Download a figure in a file format

To save a figure as a file, the README gives one route: open the slide in Google Slides and go to File, then Download, then pick the format you want. The source does not list which formats are available, so check the menu for the options it offers.

  • Open the slide containing the figure.
  • Go to File, then Download.
  • Choose the format that fits your use.
File → Download → (choose your format)

Credit the author and follow the usage terms

You do not need to ask permission to use the visuals in machine learning presentations or blog posts. The README asks that you give credit to the designer or author. Author information is stored in the slide notes, so check there before you publish.

The repository is listed under the MIT licence. Check the repository's licence file for the exact terms before reuse in a commercial or formal publication.

  • Open the slide notes section to find the author's name.
  • Cite the designer in your slide footer, caption or post.

Add your own figure or request one

To add a figure, add a new slide and reuse the basic visual components already in the deck. The README asks contributors to include author information in the slide notes and a short description of the visual, so others understand what it shows and how to use it.

If you need help customizing a figure or want a new one, open an issue on the project's GitHub issues page. The README also points to a project page and a list of figure ideas, including linear regression, backpropagation, LSTM, Transformer and attention.

  • Add a new slide and reuse existing components.
  • Put your author information in the slide notes.
  • Write a short description of the visual on the slide.
  • Open a GitHub issue to request a figure or ask for help.

What people use it for

  • Adding a neural network diagram to a machine learning lecture or thesis slide.
  • Customizing a figure for a research paper submission.
  • Illustrating a machine learning concept in a blog post.
  • Building teaching slides on topics such as neural network architectures.

Frequently asked questions

Is ML Visuals free to use?

The README describes the visuals as free and says you do not need to ask permission to use them in machine learning presentations or blog posts. It asks that you credit the designer or author. The repository is listed under the MIT licence, so check its licence file for the full terms.

How do I install ML Visuals?

There is nothing to install for the slides, which are read online in Google Slides. The README gives one command, git clone https://github.com/dair-ai/ml-visuals, if you want a local copy of the repository.

What do I need to use ML Visuals?

The README lists no prerequisites. You need a web browser to open the Google Slides, and a Google account if you want to make your own copy or request edit access.

How do I download a figure from ML Visuals?

Open the slide, then go to File, then Download, then choose your format. The README gives this route without listing the formats.

What are alternatives to ML Visuals?

The project's comparable tools are Excalidraw, draw.io, TikZ and Mermaid. Those are drawing or diagram tools that let you create your own figures rather than reuse a shared set of slides.

Is ML Visuals a stable, versioned library?

Not necessarily. The README mentions only Version 1.0, says more updates are coming, and the source shows the last push was in February 2023. If you need a fixed asset library, save a copy of the figures you rely on.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

ML Visuals suits machine learning students, researchers, bloggers and educators who want clean, reusable neural network and ML diagrams they can edit in Google Slides. It is not a software tool, and it is a poor fit if you need a stable, versioned asset library or figures on non-ML topics; for those, consider Excalidraw, draw.io, TikZ or Mermaid.

Install options & downloads