can anyone understand why reranking makes the mean reciprocal rank worse than if i do not use reranking at all. I am using a hybrid retriever (BM25 n Index).
Hi, I was trying to determine what the optimal Top k values are and I experimented with values ranging from (5-10). I noticed that hit rate increases and MRR remains quite similar when I use a higher Top K value . Does this mean that using a higher Top K value is better or do i need to consider other factors as well ?
I am trying to evaluate retrieval with reranking. I have already evaluated it without reranking and it worked, however, the code seems to be taking a long while to load when I included the node_postprocessors for reranking. I was wondering if this is the correct way to evaluate retrieval with reranking.
