Introduction to Word Embeddings

This repository is aimed to provide an abstract level introduction to different Word Embeddings in NLP.

• word_embs.ipynb file contains the code with explanation for the tutorial.

Read this repo on a blog: Semantic Representations

How to run the code

To run the code, follow these steps:

1. git clone https://github.com/panditu2015/Word-Embeddings.git
2. cd Word-Embeddings
3. pip install -r requirements.txt
4. jupyter notebook
5. Open word_embs.ipynb in your browser now.

Run on Google-Colab

To run on google-colab, follow this link:

https://colab.research.google.com/drive/1vEO_P564JAjTd-3El6C9lmZokZ0U9UnC

• From the menu on top, click Runtime
• Then click Change runtime type
• Under Hardware Accelarator, select GPU

Resources

• Most of the code is written using pymagnitude library in Python.

• Stanford's CS276: Information Retrieval and Web Search

• Stanford's CS224n: Natural Language Processing with Deep Learning

  • Useful lectures: 1 (Jan 8), 2 (Jan 10), 13 (Feb 19)

• Stanford's CS224U: Natural Language Understanding

  • Useful lectures: 2 (Apr 8), 7 (May 11)

• A really nice blog on the intuition of Word Embeddings:

  • The Illustrated Word2vec
  • Video version of this blog.

• A great blog on Contextual Embeddings:

  • The Illustrated BERT, ELMo, and co. (How NLP Cracked Transfer Learning).

• A blog on BERT:

  • A Visual Guide to Using BERT for the First Time

• Lecture on Distributed Representations from CS276.

• A lecture note on Word Embeddings from CS224n.