SiameseBERT-Notebook

Quick semantic search using Siamese-BERT encodings

The SiameseBERT-SemanticSearch.ipynb Google Colab Notebook illustrates using the Sentence Transformer python library to quickly create BERT embeddings for sentences and perform fast semantic searches.

The Sentence Transformer library is available on pypi and github. The library implements code from the ACL 2019 paper entitled "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks" by Nils Reimers and Iryna Gurevych.

To cite the original ACL paper:

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "http://arxiv.org/abs/1908.10084",
}