Comment by soVeryTired
2 years ago
It's a machine learning method for turning a sample of text into a fixed-length vector. We usually try to find embeddings with the property that samples of text with similar meanings are mapped to vectors that are close to one another.
Could you recommend a self-hosted library for playing with text embeddings (as opposed to a REST API hosted online)? Thanks in advance!
You could try SentenceTransformers: https://www.sbert.net/
Thank you!
Instructor Large and XL work great.
Why not just call it a text vector?