Text classification
This repository contains some classical models to do text classification.
Develop Requirements
- OS: Ubuntu 16.04 LTS (64bit)
- GPU: Nvidia Titan Xp
- Language: Python 3.6.2.
- Pytorch: 0.4.1
- nltk==3.2.4
- tensorboardX==0.8
- torchtext==0.3.1
Dataset
[1] TREC: TREC question dataset—task involves classifying a question into 6 question types (whether the question is about person, location, numeric information, etc.)
[2] 20newsgroup: This dataset contains messages from twenty newsgroups.
[3] MR: Movie reviews with one sentence per review. Classification involves detecting positive/negative reviews.
Models
[1] TextCNN
[2] TextRNN
[3] RCNN
Excution
python train.py -c config_cnn.json
result
(Measure: Accuracy on test data)
| Model | Dataset | MR | TREC |
|---|---|---|---|
| TextCNN(Rand) | Results | 67.7 | 88.2 |
| paper | 76.1 | 91.2 | |
| TextCNN(Static) | Results | 79.7 | 93.4 |
| paper | 81.0 | 92.8 | |
| TextCNN(Non-static) | Results | 80.1 | 90.8 |
| paper | 81.5 | 93.6 | |
| TextCNN(Multichannel) | Results | 79.8 | 92.8 |
| paper | 81.1 | 92.2 | |
| TextRNN | Results | 80.9 | 94.4 |
| paper | |||
| TextRNN+Attention | Results | 80.2 | 94.6 |
| paper | |||
| TextRCNN | Results | 80.05 | 93.2 |
| paper |
reference
[1] Text Classification Algorithms: A Survey(很不错的文本分类综述(2019))
[2] Text Classification Algorithms: A Survey(从机器学习到深度学习的文本分类综述(2019))
[3] 用深度学习（CNN RNN Attention）解决大规模文本分类问题 - 综述和实践(一位阿里人的文本分类小综述，结合了实际问题和业务)