summaryrefslogtreecommitdiffstats
path: root/searchlib/src/tests/tensor
Commit message (Expand)AuthorAgeFilesLines
* Share code for loading and saving tensor attribute betweenTor Egge2022-10-071-3/+3
* Rename search::tensor::DirectTensorStore::get_tensor to get_tensor_ptr.Tor Egge2022-10-071-4/+4
* Add unit test for TensorBufferTypeMapper.Tor Egge2022-10-072-0/+130
* Deinline TestParam copy constructor.Tor Egge2022-10-031-0/+3
* Add TensorBufferStore.Tor Egge2022-09-302-0/+173
* Add tensor buffer operations.Tor Egge2022-09-302-0/+197
* add support for multi-bitvector global filterHåvard Pettersen2022-09-131-12/+3
* GlobalFilter is now an interfaceHåvard Pettersen2022-09-121-7/+14
* fix undefined behavior in unit tests -- WIPHåvard Pettersen2022-06-071-1/+3
* Add acquire_node_ref() member function to HnswGraph.Tor Egge2022-03-221-5/+5
* Use constantHarald Musum2022-02-191-1/+1
* Somehow a paren got lost before comittingHarald Musum2022-02-181-1/+1
* Use EXPECT_NEAR when comparingHarald Musum2022-02-181-1/+1
* compare new GeoGcd with implementation as tensor distanceArne H Juul2022-01-171-0/+5
* Calculate need for compaction when updating stats.Tor Egge2021-12-091-2/+3
* Reduce alignment for small dense tensors.Tor Egge2021-12-091-1/+10
* Add CompactionSpec.Tor Egge2021-12-081-2/+6
* Move CompactionStrategy from searchcommon to vespalib.Tor Egge2021-12-081-2/+2
* Both arguments to distancecalculations will always be the same and have expec...Henning Baldersheim2021-11-261-4/+0
* Update 2020 Oath copyrights.gjoranv2021-10-272-2/+2
* Update Verizon Media copyright notices.gjoranv2021-10-077-7/+7
* Update 2017 copyright notices.gjoranv2021-10-072-2/+2
* Change loading of nearest neighbor index to use direct I/O instead of mmapping.Geir Storli2021-09-151-2/+18
* Reduce memory spike during loading of hnsw index by committing at regular int...Geir Storli2021-09-101-3/+2
* Compact HNSW index when ratio of dead bytes / address space is too highTor Egge2021-08-181-1/+117
* A swappable attribute will use a file backed memory allocator.Henning Baldersheim2021-08-121-4/+4
* Allow more rounding errors for small (zero) distances.Tor Egge2021-05-011-4/+9
* actually wants floatArne Juul2021-04-132-2/+3
* let the distance function decide how it wants the query tensor convertedArne Juul2021-04-122-2/+2
* test a bit moreArne Juul2021-04-121-1/+27
* fix NNS distance for new cell typesArne Juul2021-04-123-4/+8
* Revert "fix NNS distance for new cell types"Arnstein Ressem2021-04-123-8/+4
* fix NNS distance for new cell typesArne Juul2021-04-093-4/+8
* Remove experimental use of mmap file allocator for dense tensor data.Geir Storli2021-03-261-4/+4
* Merge pull request #16495 from vespa-engine/arnej/add-size-literalsArne H Juul2021-02-161-3/+4
|\
| * use size literals in searchlibArne Juul2021-02-151-3/+4
* | Use mmap file allocator in dense tensor attribute if huge flag isTor Egge2021-02-151-1/+2
|/
* make check_with_distance_threshold methodArne Juul2021-01-081-8/+16
* add distanceThreshold option for nearestNeighbor operatorArne Juul2021-01-082-3/+44
* move class to the only place it is usedArne Juul2020-12-101-1/+0
* getTensor with mutable view no longer neededArne Juul2020-12-031-3/+2
* use SimpleValue instead of EngineOrFactoryArne Juul2020-12-022-7/+8
* cleanup unused includeArne Juul2020-11-271-1/+0
* track CellType moveArne Juul2020-11-201-5/+5
* more robust test when tensors are memory efficientArne Juul2020-10-151-3/+28
* simplifyArne Juul2020-10-152-5/+2
* handle both engine- and factory-based tensorsArne Juul2020-10-152-25/+25
* track TypedCells moveArne Juul2020-10-013-9/+8
* memory_used -> mem_usage to reflect semantic changeArne Juul2020-09-101-9/+9
* use MemoryUsage from TensorArne Juul2020-09-091-8/+9