UER: An Open-Source Toolkit for Pre-training Models
This provides a practical toolkit for NLP researchers and practitioners to efficiently experiment with pre-training models, though it is incremental as it builds on existing methods like BERT.
The authors tackled the lack of a flexible framework for deploying various pre-training models in NLP by proposing UER, an open-source toolkit that allows users to assemble modules to reproduce or develop models, achieving new state-of-the-art results on multiple downstream datasets.
Existing works, including ELMO and BERT, have revealed the importance of pre-training for NLP tasks. While there does not exist a single pre-training model that works best in all cases, it is of necessity to develop a framework that is able to deploy various pre-training models efficiently. For this purpose, we propose an assemble-on-demand pre-training toolkit, namely Universal Encoder Representations (UER). UER is loosely coupled, and encapsulated with rich modules. By assembling modules on demand, users can either reproduce a state-of-the-art pre-training model or develop a pre-training model that remains unexplored. With UER, we have built a model zoo, which contains pre-trained models based on different corpora, encoders, and targets (objectives). With proper pre-trained models, we could achieve new state-of-the-art results on a range of downstream datasets.