Over 200 figures and diagrams of the most popular deep learning architectures and layers FREE TO USE in your blog posts, slides, presentations, or papers.

Deep Learning Visuals

Shield:

This repository was inspired by the ML Visuals repository maintained by dair.ai.

Deep Learning Visuals contains 215 unique images divided in 23 categories (some images may appear in more than one category). All the images were originally published in my book “Deep Learning with PyTorch Step-by-Step: A Beginner’s Guide”.

Can I Freely Use These Images?

Sure, these images can be FREELY USED in your own blog posts, slides, presentations, or papers under the CC-BY license.

Awesome, where are they?

You can easily navigate through the pages and indices, and click on the desired image to visualize it in full size:

• Activation Functions
• Architectures
• Assorted
• Attention
• Batch Norm
• BERT
• Classification
• Convolutions
• Decoder
• Dropout
• ELMo
• Encoder
• Feed-Forward Networks
• Gradient Descent
• Initializations and Gradient Clipping
• LayerNorm
• Optimizers and Schedulers
• Patch Embeddings
• Positional Encoding
• RNNs
• Seq2Seq
• Transformers

How Can I Use Them?

DISCLAIMER: this is NOT legal advice, you should always read the license yourself!

In a nutshell, you’re allowed to use (or adapt) these images in your own materials, even for commercial purposes, as long as you attribute it.

Here is a quick guide on Best Practices for Attribution.

Here are some examples of both images and attributions:

Logistic Regression

Image by dvgodoy / CC BY

RNN

Image by dvgodoy / CC BY

Transformer

Image by dvgodoy / CC BY

This work is licensed under a Creative Commons Attribution 4.0 International License.