Administrative Information
| Title | Convolutional Neural Networks and Transformers for images |
| Duration | 60 |
| Module | B |
| Lesson Type | Tutorial |
| Focus | Technical - Deep Learning |
| Topic | Neural Networks |
Keywords
Convolutional Neural Networks,Transformer networks,Computer Vision,
Learning Goals
- Implementing and training a CNN and a Transformer for a computer vision problem from scratch
- Comparing the performance of the two models
Expected Preparation
Learning Events to be Completed Before
Obligatory for Students
Optional for Students
None.
References and background for students
None.
Recommended for Teachers
None.
Lesson materials
Instructions for Teachers
Here the goal is to show the implementation steps of the transformer network for images. We also implement a basic CNN with a similar number of trainable parameters, so we can compare the performance of the CNN to the ViT model.
Outline
| Duration (min) | Description |
|---|---|
| 15 | Training a simple convolutional neural network |
| 35 | Training a vision transformer (ViT) neural network |
| 10 | Comparing the performance of the two different approaches |
Acknowledgements
The Human-Centered AI Masters programme was Co-Financed by the Connecting Europe Facility of the European Union Under Grant №CEF-TC-2020-1 Digital Skills 2020-EU-IA-0068.