RuntimeError when indexing with FAISS

[kalinduaet]

I'm getting this error when I try to index document with FAISS. Is there way change number or cluster?
RuntimeError: Error in void faiss::Clustering::train_encoded(faiss::idx_t, const uint8_t*, const faiss::Index*, faiss::Index&, const float*) at /project/faiss/faiss/Clustering.cpp:261: Error: 'nx >= k' failed: Number of training points (109) should be at least as large as number of clusters (128)

[alpborcek]

same for me?

[mr-shitij]

I am getting below error. any solution ?

assert classname.endswith('Vector')
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

AssertionError

[bclavie]

Hey,

• @mr-shitij, not sure what this error is, but it's an odd that is more likely to be related to your faiss install.
• @alpborcek and @kalinduaet, this should only occur when trying to create extremely small indexes. Is this what you are currently doing? Next version of RAGatouille will ship with a version that bypasses this process entirely for indexes that are small-sized, so the error will go away.

[mr-shitij]

Hey,

• @mr-shitij, not sure what this error is, but it's an odd that is more likely to be related to your faiss install.
• @alpborcek and @kalinduaet, this should only occur when trying to create extremely small indexes. Is this what you are currently doing? Next version of RAGatouille will ship with a version that bypasses this process entirely for indexes that are small-sized, so the error will go away.

Thanks for reply; It got solved there is problem in my end ..!!
Thank You.

[ergitoshkezi]

@mr-shitij Could you please tell us how did you solve this issue
Thank you

[mr-shitij]

@mr-shitij Could you please tell us how did you solve this issue Thank you

sorry I don't remember exactly, but you can try with this version "RAGatouille==0.0.8post1"