Glossary term
Glossary term
Foundations
A vector whose values are mostly zeroes. See also sparse feature and sparsity.
Created for this library
A search team stores sparse vectors of token counts as part of its hybrid retrieval index alongside dense embeddings.
An NLP team uses sparse vectors as inputs to a logistic regression baseline before evaluating embedding-based alternatives.
A research team uses sparse vectors as the input format for its bag-of-words baseline on tabular text data.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License