Background materials

CS224u has CS224n as a official prerequisite. We are assuming that you are familiar with core concepts in NLP and machine learning. The following materials may be useful to you if you need a refresher.

Basic tools

  1. Notebook: Jupyter notebook tutorial
  2. Notebook: NumPy tutorial
  3. Notebook: PyTorch tutorial
  4. Notebook: Using and extending the course PyTorch models

Static vector representations

  1. Video: High-level goals and guiding hypotheses [slides]
  2. Video: Matrix designs [slides]
  3. Video: Vector comparison [slides]
  4. Video: Basic reweighting [slides]
  5. Notebook: Designs, distances, basic reweighting
  6. Video: Dimensionality reduction [slides]
  7. Notebook: Dimensionality reduction and representation learning
  8. Notebook: Retrofitting
  9. Video: Static representations from contextual models [slides]
  10. Notebook: Static representations from contextual models

Supervised learning

  1. Tutorial videos on supervised learning
  2. Stanford AI Lab Deep Learning Tutorial
  3. Video: Overview of supervised sentiment analysis [slides]
  4. Video: General practical tips [slides]
  5. Video: Stanford Sentiment Treebank [slides]
  6. Notebook: Overview of the Stanford Sentiment Treebank
  7. Video: DynaSent [slides]
  8. Video: sst.py [slides]
  9. Video: Hyperparameter search and classifier comparison [slides]
  10. Video: Feature representation [slides]
  11. Notebook: Hand-built feature functions
  12. Video: RNN classifiers [slides]
  13. Notebook: Dense feature representations and neural networks
  14. Video: Practical fine-tuning [slides]
  15. Notebook: Fine-tuning large language models