Notebooks for the WUR course Deep Learning (AIN31306), mostly focused on learning on sequential data (i.e. text, DNA, proteins). These notebooks cover three days of a larger curriculum, which is hosted elsewhere. The notebooks are accompanied by lectures based on the book Deep Learning - Foundations and Concepts (Bishop & Bishop).
Day 1:
- Tokenization
- Word embedding
- Markov prediction
- 1D CNN
Day 2:
- 'Naive' Transformer
- GPT
Day 3:
- Prompting
- Finetune GLUEBERT