41 Commits

Author	SHA1	Message	Date
Andrew Kane	8f00d679d6	Removed type-specific code from IVFFlat - #527	2024-04-24 22:19:53 -07:00
Andrew Kane	78e5bcf229	Switched to 0-based numbering for sparsevec on-disk format	2024-04-24 12:51:24 -07:00
Andrew Kane	24c1b51099	Added comment [skip ci]	2024-04-24 10:13:50 -07:00
Heikki Linnakangas	14b351bc92	Fix integer overflow in subvector() function (#530) ... `end = start + count` can overflow if `start` is very large. That leads to a segfault later in the function. Add test case for it.	2024-04-24 01:20:16 -07:00
Andrew Kane	f9941c2992	Moved L1 distance to halfutils [skip ci]	2024-04-22 13:19:42 -07:00
Andrew Kane	ed4837cc4f	Renamed bit files	2024-04-22 10:22:18 -07:00
Andrew Kane	fb77671d05	Added casts between halfvec and sparsevec	2024-04-19 18:03:07 -07:00
Andrew Kane	55845bfd5f	Added SIMD version of cosine distance	2024-04-15 10:01:05 -07:00
Andrew Kane	38b223b4bd	Added concatenate operator for vectors	2024-04-14 23:12:07 -07:00
Andrew Kane	b70fb2b3f4	Added l2_normalize function - closes #220	2024-04-14 20:53:05 -07:00
Andrew Kane	fe9a9a0d0f	Renamed quantize_binary to binary_quantize	2024-04-14 16:57:28 -07:00
Andrew Kane	8751404a94	Renamed halfvec_dims to vector_dims [skip ci]	2024-04-14 16:40:16 -07:00
Andrew Kane	44e536b4ae	Renamed functions to l2_norm [skip ci]	2024-04-14 16:29:27 -07:00
Andrew Kane	e146f3cfb6	Added avg for half vectors [skip ci]	2024-04-14 15:11:11 -07:00
Andrew Kane	c68c2867fd	Added more functions for halfvec	2024-04-14 13:12:08 -07:00
Andrew Kane	89629abc08	Removed unneeded check [skip ci]	2024-04-13 15:06:17 -07:00
Andrew Kane	db112862a5	Improved performance of halfvec_in	2024-04-13 15:04:52 -07:00
Andrew Kane	8f93d02c71	Updated comments [skip ci]	2024-04-13 14:06:10 -07:00
Andrew Kane	c5c82bafda	Updated invalid input syntax messages to be consistent [skip ci]	2024-04-13 11:32:47 -07:00
Andrew Kane	23944302fe	Improved input functions [skip ci]	2024-04-13 11:09:59 -07:00
Andrew Kane	aed463dbad	Fixed headers	2024-04-12 11:58:27 -07:00
Andrew Kane	f64abe3aae	Fixed performance of halfvec	2024-04-12 11:50:34 -07:00
Andrew Kane	8d9400bae3	Added support for halfvec to IVFFlat	2024-04-11 19:56:39 -07:00
Andrew Kane	a4531ca51f	Fixed halfvec_cmp_internal function [skip ci]	2024-04-11 19:43:25 -07:00
Andrew Kane	e6a591275a	Added halfvec_cmp_internal function [skip ci]	2024-04-11 18:33:54 -07:00
Andrew Kane	bd52ed29e0	Added halfvec_spherical_distance function [skip ci]	2024-04-11 13:52:59 -07:00
Andrew Kane	e6ca831f3d	Store very small values as zero for halfvec	2024-04-10 13:59:51 -07:00
Andrew Kane	a824af42fb	Improved error message for out of range values for halfvec [skip ci]	2024-04-10 13:47:45 -07:00
Andrew Kane	191c8e1cca	Use consistent naming [skip ci]	2024-04-08 14:56:59 -07:00
Andrew Kane	21bcff6722	Added CPU dispatching for halfvec distance functions - #311 ... Co-authored-by: Arda Aytekin <arda.aytekin@microsoft.com>	2024-04-08 13:50:18 -07:00
Andrew Kane	60b4bb2ad1	Moved halfvec distance functions to separate file [skip ci]	2024-04-08 10:00:34 -07:00
Andrew Kane	39b8bd1816	Switched to storeu [skip ci]	2024-04-07 21:04:59 -07:00
Andrew Kane	925aa4e048	Added SIMD version of L2 distance	2024-04-07 20:22:19 -07:00
Andrew Kane	9ed39cee67	Added SIMD version of inner product	2024-04-07 20:10:54 -07:00
Andrew Kane	3bd67fef54	DRY halfvec distance functions	2024-04-07 19:03:20 -07:00
Andrew Kane	d8719d59a3	Improved halfvec performance with F16C support	2024-04-07 18:53:50 -07:00
Andrew Kane	33daf87fcb	Improved performance of HalfToFloat4	2024-04-07 13:01:58 -07:00
Andrew Kane	8961de6179	Improved halfvec input check [skip ci]	2024-04-03 22:23:23 -07:00
Andrew Kane	9f61dcff5d	Improved error messages for halfvec input [skip ci]	2024-04-03 22:20:13 -07:00
Andrew Kane	aaa2d644ce	Added quantize_binary and subvector functions for halfvec	2024-04-03 14:53:03 -07:00
Andrew Kane	32a502c838	Added halfvec type	2024-04-02 13:55:45 -07:00