The aim is to bulk index both the data and associated vectors (sentence embeddings) using sentence-transformers into Qdrant, a vector database so that we can perform similarity search on phrases.
Unlike keyword based search, similarity search requires vectors that come from an NLP (typically transformer) model
multi-qa-distilbert-cos-v1 is the model used: As per the docs, "This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs."
Unlike other cases, generating sentence embeddings on a large batch of text is quite slow on a CPU, so the aim is to explore how to generate ONNX-optimized models so that we both generate and index the vectors into db more rapidly without a GPU
Goals
The aim is to bulk index both the data and associated vectors (sentence embeddings) using
sentence-transformersinto Qdrant, a vector database so that we can perform similarity search on phrases.sentence-transformersmulti-qa-distilbert-cos-v1is the model used: As per the docs, "This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs."