liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-06-29 17:21:16 +08:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • d244a040e1 Increased max sparsevec dimensions to 1B [skip ci] Andrew Kane 2024-04-24 11:17:25 -07:00
  • c3448a25e2 Improved error messages for sparsevec input Andrew Kane 2024-04-24 11:12:28 -07:00
  • 053ce2ddae Improved CI for Windows [skip ci] Andrew Kane 2024-04-24 10:22:31 -07:00
  • 24c1b51099 Added comment [skip ci] Andrew Kane 2024-04-24 10:13:50 -07:00
  • 9696835a19 Improved tests for sparsevec input [skip ci] Andrew Kane 2024-04-24 09:58:27 -07:00
  • b2a5259607 Switched to strtoint for sparsevec input Andrew Kane 2024-04-24 09:56:09 -07:00
  • c198fd58ee Added more tests for subvector function [skip ci] Andrew Kane 2024-04-24 01:31:50 -07:00
  • 8c408759dc Added more tests for subvector function [skip ci] Andrew Kane 2024-04-24 01:28:25 -07:00
  • 14b351bc92 Fix integer overflow in subvector() function (#530) Heikki Linnakangas 2024-04-24 11:20:16 +03:00
  • ad3f811fa3 Use VARSIZE_ANY instead of itemsize to avoid uninitialized bytes Andrew Kane 2024-04-23 23:52:02 -07:00
  • 2867742f3d Test sparsevec vacuum recall debug10 Andrew Kane 2024-04-23 21:54:15 -07:00
  • 281a74f54e Improved consistency of sparsevec_l1_distance with vector [skip ci] Andrew Kane 2024-04-23 21:24:02 -07:00
  • 034713c803 Improved consistency with vector [skip ci] Andrew Kane 2024-04-23 21:13:00 -07:00
  • ed2e460f00 Improved consistency with vector [skip ci] Andrew Kane 2024-04-23 21:11:27 -07:00
  • d136615874 Improved test [skip ci] Andrew Kane 2024-04-23 20:42:30 -07:00
  • d70b160e0a Improved test [skip ci] Andrew Kane 2024-04-23 20:41:11 -07:00
  • d1affcc667 Improved tests for l2_norm [skip ci] Andrew Kane 2024-04-23 20:38:22 -07:00
  • 158481ff2a Improved tests for sparsevec distance functions [skip ci] Andrew Kane 2024-04-23 20:29:04 -07:00
  • 794bbaecc7 Removed padding for sparsevec - #529 Andrew Kane 2024-04-23 20:07:24 -07:00
  • 8eddcfbd1d Increased max sparsevec dimensions to 1M [skip ci] Andrew Kane 2024-04-23 17:47:11 -07:00
  • 8b0443afdf Test sparsevec vacuum recall valgrind-sparsevec-vacuum Andrew Kane 2024-04-23 17:10:48 -07:00
  • b609c343b4 Moved type-specific code to separate functions Andrew Kane 2024-04-23 16:32:10 -07:00
  • bbfb3f200a DRY code for sorting vector arrays [skip ci] Andrew Kane 2024-04-23 15:59:42 -07:00
  • 99d367edc0 Improved code [skip ci] Andrew Kane 2024-04-23 15:53:12 -07:00
  • 991743786a Set length for newCenters and aggCenters [skip ci] Andrew Kane 2024-04-23 15:47:04 -07:00
  • 60ceaea4f2 Added safety check to NormCenters [skip ci] Andrew Kane 2024-04-23 15:43:04 -07:00
  • 9cd789fe06 Switched to support function for normalizing centers for k-means Andrew Kane 2024-04-23 15:39:58 -07:00
  • f26bd7054d Moved varbit check [skip ci] index-type-support-v2 Andrew Kane 2024-04-23 14:22:05 -07:00
  • 4305e723cc Fixed arguments [skip ci] Andrew Kane 2024-04-23 14:06:28 -07:00
  • 571a277af9 Removed type-specific code from HNSW [skip ci] Andrew Kane 2024-04-23 13:54:04 -07:00
  • 0da6213a60 Moved type lookup to support functions - #527 Andrew Kane 2024-04-23 13:02:47 -07:00
  • 3172d7ef57 Moved type support to support functions [skip ci] index-type-support Andrew Kane 2024-04-23 12:51:04 -07:00
  • d1b83991af Forbid zero values in sparsevec's binary input function (#528) Heikki Linnakangas 2024-04-23 19:13:53 +03:00
  • 6c247a38d3 Updated readme [skip ci] Andrew Kane 2024-04-22 21:56:05 -07:00
  • 6639cde19d Updated readme [skip ci] Andrew Kane 2024-04-22 21:32:59 -07:00
  • bd409f0c6a Moved HnswGetType call [skip ci] Andrew Kane 2024-04-22 19:22:09 -07:00
  • 1994fd003a Removed unneeded headers [skip ci] Andrew Kane 2024-04-22 19:10:50 -07:00
  • bd62561a19 Added support function for l2_normalize to ivfflat Andrew Kane 2024-04-22 19:06:06 -07:00
  • f14c21748b Added support function for l2_normalize [skip ci] Andrew Kane 2024-04-22 18:36:47 -07:00
  • 2b77005610 Removed type-specific code from ivfscan Andrew Kane 2024-04-22 18:12:18 -07:00
  • e884b3aa69 Added opclasses to readme [skip ci] Andrew Kane 2024-04-22 16:29:34 -07:00
  • ab71c12a28 Added comments on dispatching [skip ci] Andrew Kane 2024-04-22 16:18:57 -07:00
  • 1804c63e27 Added more tests for vector distance functions [skip ci] Andrew Kane 2024-04-22 15:53:13 -07:00
  • 4e6aa2f0c1 Added DISABLE_DISPATCH option [skip ci] Andrew Kane 2024-04-22 15:43:07 -07:00
  • 40e86251c3 Added VECTOR_TARGET_CLONES to VectorL1Distance [skip ci] Andrew Kane 2024-04-22 15:15:57 -07:00
  • 0c9ae4b187 Added CPU dispatching for L1 distance for halfvec Andrew Kane 2024-04-22 15:02:17 -07:00
  • d83af48e70 Improved tests for halfvec l1_distance [skip ci] Andrew Kane 2024-04-22 14:43:54 -07:00
  • b2f7dad8a7 Removed support for L1 distance and Jaccard distance from ivfflat due to non-optimal clustering Andrew Kane 2024-04-22 14:11:29 -07:00
  • 881fbc15ef Added L1 distance operator to docs [skip ci] Andrew Kane 2024-04-22 13:22:28 -07:00
  • f9941c2992 Moved L1 distance to halfutils [skip ci] Andrew Kane 2024-04-22 13:19:42 -07:00
  • f9c071a761 Improved tests for L1 distance with halfvec Andrew Kane 2024-04-22 13:14:45 -07:00
  • 9f4b770db3 Added support for indexing sparsevec with L1 distance [skip ci] Andrew Kane 2024-04-22 13:08:12 -07:00
  • 70b299a7ff Added support for indexing halfvec with L1 distance [skip ci] Andrew Kane 2024-04-22 13:00:59 -07:00
  • d46d014234 Updated test name [skip ci] Andrew Kane 2024-04-22 12:57:01 -07:00
  • 655adc535a Added bit examples for ivfflat [skip ci] Andrew Kane 2024-04-22 12:51:38 -07:00
  • 47f0a5e9ac Updated test name [skip ci] Andrew Kane 2024-04-22 12:48:22 -07:00
  • af9d50481d Added support for indexing L1 distance Andrew Kane 2024-04-22 12:44:03 -07:00
  • 6dad8460a6 Updated readme [skip ci] Andrew Kane 2024-04-22 10:45:41 -07:00
  • ed4837cc4f Renamed bit files Andrew Kane 2024-04-22 10:22:18 -07:00
  • 3df5655b30 Removed unneeded headers [skip ci] Andrew Kane 2024-04-19 18:09:38 -07:00
  • e4c4ac9648 Added more tests for sparsevec to halfvec cast [skip ci] Andrew Kane 2024-04-19 18:05:03 -07:00
  • fb77671d05 Added casts between halfvec and sparsevec Andrew Kane 2024-04-19 18:03:07 -07:00
  • fd4fbd238c Updated sparsevec input to support indices in any order [skip ci] Andrew Kane 2024-04-19 16:54:19 -07:00
  • 4e093f95be Improved consistency of CPU dispatching code Andrew Kane 2024-04-18 14:13:12 -07:00
  • fb3c964ac2 Improved performance of bit distance functions - #519 Andrew Kane 2024-04-18 13:45:00 -07:00
  • 0b938f8328 Improved halfvec dispatching code [skip ci] Andrew Kane 2024-04-17 20:22:14 -07:00
  • 6153e173f3 Updated comments [skip ci] Andrew Kane 2024-04-17 18:57:59 -07:00
  • eb48b9eec4 Added comments [skip ci] Andrew Kane 2024-04-17 18:53:58 -07:00
  • fb6af03361 Fixed CPU dispatching check for halfvec distance functions Andrew Kane 2024-04-17 18:27:27 -07:00
  • e9d5a0b1aa Improved check [skip ci] half-dispatch Andrew Kane 2024-04-17 18:19:03 -07:00
  • 52a81b3b9d Improved check [skip ci] Andrew Kane 2024-04-17 18:16:58 -07:00
  • c5372e493b Improved check [skip ci] Andrew Kane 2024-04-17 18:13:58 -07:00
  • e2d8907180 Added todo [skip ci] Andrew Kane 2024-04-17 17:25:53 -07:00
  • 09ea1b0d5c Improved target_clones logic [skip ci] Andrew Kane 2024-04-17 15:05:47 -07:00
  • 8d68f88507 Improved target_clones logic [skip ci] Andrew Kane 2024-04-17 15:05:19 -07:00
  • 301c8083f5 Added check for undefined behavior to CI [skip ci] Andrew Kane 2024-04-17 02:03:46 -07:00
  • 8b33a359ce Updated VECTOR_ARRAY_SIZE for alignment [skip ci] Andrew Kane 2024-04-17 00:55:28 -07:00
  • 77ec24641e Fixed flaky test [skip ci] Andrew Kane 2024-04-17 00:27:37 -07:00
  • 576a37e975 Ensure items are always aligned Andrew Kane 2024-04-17 00:17:40 -07:00
  • cf570810eb Fixed alignment for bit in IVFFlat - fixes #520 Andrew Kane 2024-04-17 00:05:31 -07:00
  • c361f80465 Synced recall test [skip ci] Andrew Kane 2024-04-16 17:27:06 -07:00
  • e730fef99d Fixed flaky test [skip ci] Andrew Kane 2024-04-16 17:25:50 -07:00
  • 4e2b76e627 Skip duplicate center check for bit [skip ci] Andrew Kane 2024-04-16 17:20:48 -07:00
  • 04af15c9d6 Added support for bit to IVFFlat Andrew Kane 2024-04-16 17:12:27 -07:00
  • 819b6cf312 Added comments to vector.sql [skip ci] Andrew Kane 2024-04-16 15:47:25 -07:00
  • 2dc021d935 Added category/preferred to types for new installations [skip ci] type-category Andrew Kane 2024-04-16 13:26:03 -07:00
  • 31dfd3d1a6 Added type to assertion message [skip ci] Andrew Kane 2024-04-16 13:11:37 -07:00
  • 8df8dd01b9 Added halfvec to distance functions TAP test Andrew Kane 2024-04-16 13:09:44 -07:00
  • b9b30cc16e Added function name to assertion message [skip ci] Andrew Kane 2024-04-16 12:20:22 -07:00
  • e1565af0dc Added TAP test for sparsevec distance functions Andrew Kane 2024-04-16 12:11:54 -07:00
  • 26d6fbb1d0 Updated comment [skip ci] Andrew Kane 2024-04-16 11:45:19 -07:00
  • 6c61b5ce6f Updated comment [skip ci] Andrew Kane 2024-04-16 11:45:03 -07:00
  • 588873e145 Added comment [skip ci] Andrew Kane 2024-04-16 11:38:59 -07:00
  • 7adb8a6e0e Fixed sparsevec comparison function and added test for consistency Andrew Kane 2024-04-16 11:36:45 -07:00
  • 5b0eff9dae Updated comment [skip ci] Andrew Kane 2024-04-15 15:53:07 -07:00
  • 5cf75c0b83 Added comment [skip ci] Andrew Kane 2024-04-15 15:50:59 -07:00
  • 5215c28923 Moved norm check to separate function Andrew Kane 2024-04-15 15:32:08 -07:00
  • 342d82be65 Improved l2_normalize test for sparsevec [skip ci] Andrew Kane 2024-04-15 14:59:11 -07:00
  • bf3ef4de56 Improved safety check [skip ci] Andrew Kane 2024-04-15 14:55:32 -07:00
  • 2c82f05503 Improved safety check [skip ci] Andrew Kane 2024-04-15 14:51:24 -07:00