Understanding the VectorStoreIndex
As this ebook edition doesn't have fixed pagination, the page numbers below are hyperlinked for reference only, based on the printed edition of this book.
In LlamaIndex, the VectorStoreIndex stands out as the workhorse, being the most commonly utilized type of Index.
For most RAG applications, a VectorStoreIndex might be the best solution because it facilitates the construction of Indexes on collections of Documents where embeddings for the input text chunks are stored within the Vector Store of the Index. Once constructed, this Index can be used for efficient querying because it allows for similarity searches over the embedded representations of the text, making it highly suitable for applications requiring fast retrieval of relevant information from a large collection of data. Don’t worry if you’re not yet familiar with terms such as embeddings, vector store, or similarity searching, because we’ll cover them in the following...
