Skip to main content
Join the official 2019 Python Developers SurveyStart the survey!

Multi-class text categorization using state-of-the-art pre-trained contextualized language models.

Project description

Text2Class

Build multi-class text classifiers using state-of-the-art pre-trained contextualized language models, e.g. BERT. Only a few hundred samples per class are necessary to get started.

Background

This project is based on our study: Transfer Learning Robustness in Multi-Class Categorization by Fine-Tuning Pre-Trained Contextualized Language Models.

Citation

To cite this work, use the following BibTeX citation.

@article{transfer2019multiclass,
  title={Transfer Learning Robustness in Multi-Class Categorization by Fine-Tuning Pre-Trained Contextualized Language Models},
  author={Liu, Xinyi and Wangperawong, Artit},
  journal={arXiv preprint arXiv:1909.03564},
  year={2019}
}

Installation

pip install text2class

Example usage

Create a dataframe with two columns, 'text' and 'label'. No text pre-processing is necessary.

import pandas as pd
from text2class import TextClassifier

df = pd.read_csv("data.csv")

train = df.sample(frac=0.9,random_state=200)
test=df.drop(train.index)

cls = TextClassifier(
	num_labels=3,
	data_column="text",
	label_column="label",
	max_seq_length=128
)

cls.fit(train)

predictions = cls.predict(test)

Advanced usage

Model type

The default model is an uncased Bidirectional Encoder Representations from Transformers (BERT) consisting of 12 transformer layers, 12 self-attention heads per layer, and a hidden size of 768. Below are all models currently supported that you can specify with hub_module_handle. We expect that more will be added in the future. For more information, see BERT's GitHub.

https://tfhub.dev/google/bert_uncased_L-12_H-768_A-12/1
https://tfhub.dev/google/bert_uncased_L-24_H-1024_A-16/1
https://tfhub.dev/google/bert_cased_L-12_H-768_A-12/1
https://tfhub.dev/google/bert_cased_L-24_H-1024_A-16/1
https://tfhub.dev/google/bert_chinese_L-12_H-768_A-12/1
https://tfhub.dev/google/bert_multi_cased_L-12_H-768_A-12/1

cls = TextClassifier(
	num_labels=3,
	data_column="text",
	label_column="label",
	max_seq_length=128,
	hub_module_handle="https://tfhub.dev/google/bert_uncased_L-12_H-768_A-12/1"
)

Contributing

Text2Class is an open-source project founded and maintained to better serve the machine learning and data science community. Please feel free to submit pull requests to contribute to the project. By participating, you are expected to adhere to Text2Class's code of conduct.

Questions?

For questions or help using Text2Class, please submit a GitHub issue.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for text2class, version 0.0.2
Filename, size File type Python version Upload date Hashes
Filename, size text2class-0.0.2-py3-none-any.whl (37.5 kB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size text2class-0.0.2.tar.gz (34.6 kB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page