Its amazingly powerful
@ The fifth elephant MLOps conference
Make work from home fun once again
Encoding text longer than 512 tokens
Making better domain text encoders with TSDAE
Yes, this might be the one
Years of sweating and burning fingers
NLP for startup engineers
Cosine similarity on a subset of documents for multipass searchHow do you calculate cosine on a subset of vectors in a vector index? Documents 1M -> 10000 -> 100 -> 10 64D -> 128D -> 256D -> 512D (D - vector…
Minimum viable reading
Better search for Indian language data
The world’s most valuable resource is no longer oil, but data.