https://github.com/kyubyong/nlp_made_easy

Explains nlp building blocks in a simple manner.
https://github.com/kyubyong/nlp_made_easy

beam-search bpe nlp seq2seq

Last synced: about 1 month ago
JSON representation

Explains nlp building blocks in a simple manner.

• Host: GitHub
• URL: https://github.com/kyubyong/nlp_made_easy
• Owner: Kyubyong
• Created: 2019-01-18T00:42:53.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2019-09-23T01:27:39.000Z (over 5 years ago)
• Last Synced: 2024-12-14T02:41:48.793Z (about 2 months ago)
• Topics: beam-search, bpe, nlp, seq2seq
• Language: Jupyter Notebook
• Homepage:
• Size: 278 KB
• Stars: 251
• Watchers: 14
• Forks: 36
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# NLP Made Easy

Simple code notes for explaining NLP building blocks

* [Subword Segmentation Techniques](Subword%20Segmentation%20Techniques.ipynb)

  * Let's compare various tokenizers, i.e., nltk, BPE, SentencePiece, and Bert tokenizer.

* [Beam Decoding](Beam%20Decoding.ipynb)

  * Beam decoding is essential for seq2seq tasks. But it's notoriously complicated to implement. Here's a relatively easy one, batchfying candidates.

* [How to get the last hidden vector of rnns properly](How%20to%20get%20the%20last%20hidden%20vector%20of%20rnns%20properly.ipynb)

  * We'll see how to get the last hidden states of Rnns in Tensorflow and PyTorch.

* [Tensorflow seq2seq template based on the g2p task](Tensorflow%20seq2seq%20template%20based%20on%20g2p.ipynb)

  * We'll write a simple template for seq2seq using Tensorflow. For demonstration, we attack the g2p task. G2p is a task of converting graphemes (spelling) to phonemes (pronunciation). It's a very good source for this purpose as it's simple enough for you to up and run.

* [PyTorch seq2seq template based on the g2p task](PyTorch%20seq2seq%20template%20based%20on%20the%20g2p%20task.ipynb)

  * We'll write a simple template for seq2seq using PyTorch. For demonstration, we attack the g2p task. G2p is a task of converting graphemes (spelling) to phonemes (pronunciation). It's a very good source for this purpose as it's simple enough for you to up and run.

* [Attention mechanism](Work in progress)

* [POS-tagging with BERT Fine-tuning](Pos-tagging%20with%20Bert%20Fine-tuning.ipynb)

  * BERT is known to be good at Sequence tagging tasks like Named Entity Recognition. Let's see if it's true for POS-tagging.

* [Dropout in a minute](Dropout%20in%20a%20minute.ipynb)

  * Dropout is arguably the most popular regularization technique in deep learning. Let's check again how it work.

* Ngram LM vs. rnnlm(WIP)

* [Data Augmentation for Quora Question Pairs](Data%20Augmentation%20for%20Quora%20Question%20Pairs.ipynb)

  * Let's see if it's effective to augment training data in the task of quora question pairs.