I have about a million documents indexed with a dense vector. I am trying to add a new field to each document with a slice of its vector, e.g. the first 700 elements only. What is the best way to do this please?
The reason is so I can use the ANN search functionality with the truncated vectors.
If you just going to use the new field in your searches, the best way is to create a new index with a new dense_vector field, and index modified vectors into it. The modification you can do on your application side.
If you are going to use both fields: new and old, then you need to reindex the whole index. You can use script to modify the content of your vectors.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.