liveBook · Manning

Vision Transformer

As explained in Inside AI

A vision transformer is a transformer-based model that processes image representations for computer-vision tasks.

A vision transformer is a transformer-based model for computer vision. It can serve as an alternative to convolutional neural networks when classifying images such as photographs of cats and dogs. section 3.4

The model receives images represented numerically: each pixel becomes three values describing its red, green, and blue components. Training adjusts neural-network weights using labeled examples, while separate testing data checks performance on unseen images. section 3.4 section 3.4

Classifying cats and dogs

A practical project trains a vision transformer on color photographs labeled as cats or dogs. High-quality training data supports learning, and held-out photographs reveal whether the model generalizes beyond its training examples. section 3.4 section 3.4

How Inside AI covers it

How deep it goes
Defines image encoding and training evaluation, but does not explain transformer internals or architecture details.
Examples
Uses RGB photographs of cats and dogs, labeled training data, and held-out testing images.
What it assumes
Assumes basic familiarity with deep learning and neural networks while introducing pixel-based RGB image representation.
What you can do after
Evaluate image-classification models by separating labeled training photographs from previously unseen testing photographs.
How it connects
Connects vision transformers with deep learning, neural networks, convolutional neural networks, supervised learning, and image classification.
Cover of Inside AI
liveBookmanning.com
Read Inside AI on liveBook Get the book at manning.com

Also explained in

Compare how 7 books explain Vision Transformer →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Inside AI. The section links point to the parts of the book it draws on.