In this video, You will learn how to implement vision transformer architecture for satellite image classification step by step in the simplest way and from scratch. In this video, we use the EuroSat benchmark dataset for satellite image classification using vision transformer. We will explain how to prepare the dataset and how to use it for this task.

Vision Transformer or ViT is a transformer model introduced by Google Brain in 2020. This novel model uses an attention mechanism particularly the selfattention mechanism in its architecture.

Don't forget to subscribe to our channel for more informative tutorials!

The EuroSat dataset can be found in the following link:
https://github.com/phelber/eurosat#

The code examples are available on our GitHub page:
https://github.com/BEEILAB/Attention...

Subscribe for more tutorials like this:    / @beeilabtv  

RELATED VIDEOS
Playlist with all our videos on attention mechanism in geospatial science:
   • Attention Mechanism: What is Attentio...  

Other Social:
Instagram:   / beeilab  
Website: https://mahdianpari.ca/
Twitter: https://twitter.com/beeilab?lang=en
LinkedIn:   / beeilab  

0:39 EuroSat Benchmark Dataset Download
1:17 Data Preparation
7:02 Vision Transformer Architecture
8:15 Vision Transformer Implementation in Tensorflow