sense2vec: Contextually-keyed word vectors

sense2vec (Trask et. al, 2015) is a nice twist on word2vec that lets you learn more interesting and detailed word vectors. This library is a simple Python implementation for loading, querying and training sense2vec models. For more details, check out our blog post. To explore the semantic similarities across all Reddit comments of 2015 and 2019, see the interactive demo.

Version 2.0 (for spaCy v3) out now! Read the release notes here.

Azure Pipelines Current Release Version pypi Version Code style: black



To finish reading, please visit source site