https://github.com/longxingtan/text-classification
Text classification in Tensorflow1.x
https://github.com/longxingtan/text-classification
attention bert capsule-network nlp python tensorflow text-classification
Last synced: 6 months ago
JSON representation
Text classification in Tensorflow1.x
- Host: GitHub
- URL: https://github.com/longxingtan/text-classification
- Owner: LongxingTan
- Created: 2019-01-02T08:41:00.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2022-10-24T01:50:19.000Z (almost 3 years ago)
- Last Synced: 2025-03-28T13:21:16.022Z (6 months ago)
- Topics: attention, bert, capsule-network, nlp, python, tensorflow, text-classification
- Language: Python
- Homepage:
- Size: 1.42 MB
- Stars: 24
- Watchers: 1
- Forks: 10
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Text_classification
## Overview
The repository implements the common algorithms for multi-class text classification.
Note that it's just prototypes for experimental purposes only
- Word or char level representation: chi-square+tfidf, word2vec, glove, fasttext, elmo, bert, or concated one
- Model: CNN, BiLSTM, Self-attention,C-LSTM, RCNN, Capsule, HAN, SVM, XGBoost
- Multi task learning: for more than one multi_labels
## Dependencies
```pip install -r requirements.txt```
- Python 3.6
- Tensorflow 1.12.0
## Usage
`python run_classifier.py`
- in config.py, set the `new_data=True`, -> generate the `./data/*.tf_record` -> utilize config.py parameters
- in config.py, set the `new_data=False`, -> utilize the data from `./data/*.tf_record` -> utilize config.json parameters
## Pretrained
- [word2vec Chinese pretrained download](https://github.com/Embedding/Chinese-Word-Vectors)
- [fasttext Chinese pretrained download](https://fasttext.cc/docs/en/crawl-vectors.html)
- [bert Chinese pretrained download from google](https://storage.googleapis.com/bert_models/2018_11_03/chinese_L-12_H-768_A-12.zip)
- tips: make sure the text use the similar preprocessing trick like segmentation as the pretrained material
- create a word2vec pretrained model [reference](https://github.com/LongxingTan/Machine_learning_python/tree/master/nlp_embedding)
## Purpose
- The classification is used to clarify the damaged part and damage type from vehicles comments
## Evaluation
- Check in tensorboard: `tensorboard --logdir=./outputs`
- Due to we have too many categories of labels (ca. 500 class for 100,000 examples), and they are not equally important, so we don’t use Macro- evaluation. And the Micro- precision/recall/F1 is the same for multi-label classification. So we check the accuracy and weighted F1.
## Ignored property
- Sometimes in one sample, more than one label are valid
- Some labels have hierarchy relationship
- imbalance issue: weighted loss, data argument, anomaly detection, upsampling and downsampling
## Todo
- Multi-task learning
- Multi-label classification