Natural Language Processing Projects

Overview

This is where I play with NLP, aka Natural Language Processing, the art of teaching machine to understand and mimic human's natural language (semantically and/or syntactically).

Projects in this repository are still in progress and gradually updated. README's content will be updated accordingly. Maybe I'll write some blog posts if necessary.

Content

At present, this repository contains the projects below.

1/ Embeddings

DONE:

• CBOW
• skip-gram

TODO:

• Rewrite in Tensorflow 2.0
• Add Pytorch implementation
• Write blog posts
• Add more types of embeddings such as GloVe, etc

2/ Text Generator

DONE:

• Tensorflow implementation (both in 1.x and 2.0)
• Blog post for Tensorflow implementation: link
• Pytorch implementation
• Blog post for Pytorch implementation: link

3/ Machine Translation

DONE:

• Tensorflow implementation (no-attention/Bahdanau/Luong)
• Blog posts for Tensorflow implementation:
  • Data Preparation: link
  • Model Creation: link
  • Training: link

TODO:

• Add Pytorch implementation
• Blog post for Pytorch implementation
• Rewrite in Tensorflow 2.0
• Blog post for Tensorflow 2.0 implementation

4/ Chat Bot

Done:

• Tensorflow implementation

TODO:

• Rewrite in Tensorflow 2.0
• Add Pytorch implementation
• Write blog posts

5/ Sentiment Analysis

Done:

• Tensorflow 2.0 implementation for:
  • IMDB Movie Reviews Classification
  • Quora Insincere Questions Classification
• Blog post for IMDB:
  • English version: link
  • Japanese version: link

TODO:

• Add Pytorch implementation
• Write blog post for Quora data

6/ POS tagging

Done:

• Pytorch implementation for toy example

TODO:

• Update Pytorch implementation for real dataset
• Add Tensorflow 2.0 implementation
• Write blog posts