--- language: - en inference: false datasets: - conll2003 - wnut_17 - jnlpba - conll2012 - BTC tags: - PyTorch --- # BERT base uncased model pre-trained on 5 NER datasets Model was trained by _SberIDP_. The pretraining process and technical details are described [in this article](https://habr.com/ru/company/sberbank/blog/). * Task: `NER` * Training Data is 5 datasets: [CoNLL-2003](https://aclanthology.org/W03-0419.pdf), [WNUT17](http://noisy-text.github.io/2017/emerging-rare-entities.html), [JNLPBA](http://www.geniaproject.org/shared-tasks/bionlp-jnlpba-shared-task-2004), [CoNLL-2012 (OntoNotes)](https://aclanthology.org/W12-4501.pdf), [BTC](https://www.derczynski.com/papers/btc.pdf) The model is pretrained for NER task using [Reptile](https://openai.com/blog/reptile/) and can be finetuned for new entities with only a small amount of samples.