Small-Text: Active Learning for Text Classification in Python

TL;DR

small-text is a versatile Python library that simplifies active learning for text classification, supporting various classifiers, strategies, and GPU acceleration, and enabling efficient experimentation and application development.

Contribution

It introduces small-text, a comprehensive active learning library integrating multiple ML frameworks and strategies, facilitating rapid development and evaluation.

Findings

SetFit matches transformer accuracy in classification

SetFit outperforms in AUC metrics

Library supports GPU-accelerated query strategies

Abstract

We introduce small-text, an easy-to-use active learning library, which offers pool-based active learning for single- and multi-label text classification in Python. It features numerous pre-implemented state-of-the-art query strategies, including some that leverage the GPU. Standardized interfaces allow the combination of a variety of classifiers, query strategies, and stopping criteria, facilitating a quick mix and match, and enabling a rapid and convenient development of both active learning experiments and applications. With the objective of making various classifiers and query strategies accessible for active learning, small-text integrates several well-known machine learning libraries, namely scikit-learn, PyTorch, and Hugging Face transformers. The latter integrations are optionally installable extensions, so GPUs can be used but are not required. Using this new library, we investigate the performance of the recently published SetFit training paradigm, which we compare to vanilla transformer fine-tuning, finding that it matches the latter in classification accuracy while outperforming it in area under the curve. The library is available under the MIT License at https://github.com/webis-de/small-text, in version 1.3.0 at the time of writing.