Skip to content

Simple implementations of NLP models. Tutorials are written in Chinese on my website https://mofanpy.com

License

Notifications You must be signed in to change notification settings

MorvanZhou/NLP-Tutorials

Repository files navigation

Natural Language Processing Tutorial

Tutorial in Chinese can be found in mofanpy.com.

This repo includes many simple implementations of models in Neural Language Processing (NLP).

All code implementations in this tutorial are organized as following:

  1. Search Engine
  1. Understand Word (W2V)
  1. Understand Sentence (Seq2Seq)
  1. All about Attention
  1. Pretrained Models

Thanks for the contribution made by @W1Fl with a simplified keras codes in simple_realize. And the a pytorch version of this NLP tutorial made by @ruifanxu.

Installation

$ git clone https://github.com/MorvanZhou/NLP-Tutorials
$ cd NLP-Tutorials/
$ sudo pip3 install -r requirements.txt

TF-IDF

TF-IDF numpy code

TF-IDF short sklearn code

image

Word2Vec

Efficient Estimation of Word Representations in Vector Space

Skip-Gram code

CBOW code

image image image

Seq2Seq

Sequence to Sequence Learning with Neural Networks

Seq2Seq code

image

CNNLanguageModel

Convolutional Neural Networks for Sentence Classification

CNN language model code

image

Seq2SeqAttention

Effective Approaches to Attention-based Neural Machine Translation

Seq2Seq Attention code

image image

Transformer

Attention Is All You Need

Transformer code

image image image

ELMO

Deep contextualized word representations

ELMO code

image image

GPT

Improving Language Understanding by Generative Pre-Training

GPT code

image image

BERT

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

BERT code

My new attempt Bert with window mask

image image

About

Simple implementations of NLP models. Tutorials are written in Chinese on my website https://mofanpy.com

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages