Jianshu_Zhao

Hello James, I will start a new issue. Another question to you: is annoy robust to high dimension dataset (or more precisely, high local intrinsic dimension dataset with LID>20, see...

Hello James, I want to revisit this topic. As I said before, after building the hnsw, k best neighbors can be extracted from hnsw graph bottom layer (layer 0) in...

Hello James, I am not sure about the annoy big O. They refuse to provide one. I just analyze based on my understandings of binary search tree, which is very...

A quick answer, with k=50, hnsw is about 2 times faster than annoy for now, for the fashion dataset, embedding results are nearly the same (small differences). I will attached...

I am trying using the largest dataset I have even seen, the Higgs boson dataset generated by large hadron collider: https://archive.ics.uci.edu/ml/datasets/HIGGS You can download the data folder and the HIGGS.csv...

Hello, Can you provide an example how to prepare a build.rs only for MacOS? I saw that both ndarray and ndarray-linalg rely directly on openblas-src, intel-mkl-src et.al. Thanks, Jianshu

Hello all, Please help if you can, this crate: https://github.com/jean-pierreBoth/annembed I tried your suggestion and use blas-src, accelerate as the backend, but it does not work, the 3 features, openblas-system...

Hello @matthagan15 , I tried but it did not work because openblas is still linked. ndarray does rely on blas-src. So the problem is with openblas but not blas-src. This...

With the above configuration, accelerate is actually working, but openblas is still linked, see my binary link status after using the openblas-system feature (via otool -L on macOS): /System/Library/Frameworks/Accelerate.framework/Versions/A/Accelerate (compatibility...

That is the openblas-system feature in ndarray-linalg is still using openblas-src system feature. Any idea why? Thanks, Jianshu