hnswlib
hnswlib copied to clipboard
Header-only C++/python library for fast approximate nearest neighbors
This is a bit of a big one - happy to break this into smaller PRs if that would be useful. This pull request: - Adds new Python bindings for...
Built ok a few months ago, however now.. cc1plus: error: bad value (‘tigerlake’) for ‘-march=’ switch ``` Building wheel for hnswlib (PEP 517) ... error ERROR: Command errored out with...
Hi Team, I am getting RuntimeError: Not enough memory when my max_corpus_size = 2595491. Could you please let me know if any solution for this. Thanks, Siva Raghuram Susarla.
Two questions regarding hnswlib's normalized vectors when using cosine: **Question 1:** When I benchmark my hnswlib indices for recall, I'm getting near-zero scores -- values like 3.41130717333766e-08. The way I'm...
Hi Thanks for the great job and implementation. Let's assume my world has N vectors. I have noticed that: 1. if I run search where k=N (get me all vectors),...
Hello and thank you for the amazing library I'm widely using in my projects! Do you have any plans of adding L1 as one of the distance functions? I know...
Hi, would it be possible to somehow get the list of data points from the first layer of the HNSW graph? Context: 1. I'm storing HNSW graphs on disk in...
This WIP PR, context in #342
if i didnt't use "set_num_threads", hnsw index would use default threads values which i printed(num_threads) were 96, why the default num_threads were 96? is it set according to the system?
i build hnsw index ( cosine distance ) based on 1000w data, however, the recall performance is bad, some query data (cosine distance very small ) not be recalled, while...