Closed John42506176Linux closed 2 months ago
Hello,
It seems that the embeddings you are feeding to the rerank function does not have a correct shape. You have a boilerplate here that shows how to encode your queries and documents to work with rerank.
Could you check if your code corresponds and if not, send the code to reproduce the issue.
Hi @NohTow,
Thanks for the assistance, I didn't realize that rerank and the retrieval had different syntaxes for passing documents, my mistake. :)
Pytorch version: Version: 2.4.1
Full Error: Code: reranked_documents = rank.rerank( documents_ids=ids, queries_embeddings=queries_embeddings, documents_embeddings=documents_embeddings, ) RuntimeError: einsum(): the number of subscripts in the equation (3) does not match the number of dimensions (2) for operand 1 and no ellipsis was given