summaryrefslogtreecommitdiffstats
path: root/searchlib
Commit message (Expand)AuthorAgeFilesLines
* Use a context for transform stateJon Bratseth2018-01-106-51/+73
* Merge pull request #4589 from vespa-engine/bratseth/integrate-tensorflowJon Bratseth2018-01-101-1/+1
|\
| * Basic TensorFlow integrationJon Bratseth2018-01-091-1/+1
* | Cache the end iterator as it now is created on the fly.Henning Baldersheim2018-01-092-6/+14
* | - Be explicit about std::make_uniqueHenning Baldersheim2018-01-091-8/+4
* | Use nested namespaceHenning Baldersheim2018-01-092-31/+17
* | Stick with a LockGuard.Henning Baldersheim2018-01-081-2/+4
* | Do explicit shutdown of executors early.Henning Baldersheim2018-01-082-2/+5
* | - Remove some unused code.Henning Baldersheim2018-01-083-113/+39
* | Use commit executor as that is the writer.Henning Baldersheim2018-01-081-1/+1
|/
* Let implementation be visible too.Henning Baldersheim2017-12-293-0/+3
* Make constructor explicit.Henning Baldersheim2017-12-282-3/+2
* Merge pull request #4499 from vespa-engine/bratseth/long-dimensionsLester Solbakken2017-12-213-8/+6
|\
| * Use longs for dimensions lengths in all API'sJon Bratseth2017-12-193-8/+6
* | SimplifyJon Bratseth2017-12-201-14/+5
* | Only import onceJon Bratseth2017-12-203-45/+47
* | SimplifyJon Bratseth2017-12-202-25/+21
* | Model signatures in import resultsJon Bratseth2017-12-204-61/+123
|/
* Ignore test needing JNI codeJon Bratseth2017-12-181-0/+2
* Revert "Revert "Bratseth/tensorflow models""Jon Bratseth2017-12-1822-59/+5832
* Revert "Bratseth/tensorflow models"gjoranv2017-12-1722-5832/+59
* Revert "Disable test relying on JNI code"gjoranv2017-12-171-2/+0
* Disable test relying on JNI codeJon Bratseth2017-12-171-0/+2
* Remove warningJon Bratseth2017-12-131-1/+1
* Use the right protobuf version againJon Bratseth2017-12-131-0/+1
* Merge with masterJon Bratseth2017-12-1324-121/+112
|\
| * Use standard locking in searchlib (pass 2).Tor Egge2017-12-124-15/+14
| * Eliminate mutated version of Runnable.Tor Egge2017-12-124-61/+17
| * Use standard locking in searchlib (pass 1).Tor Egge2017-12-128-40/+41
| * Factor out CountDownLatch and Gate to separate header files.Tor Egge2017-12-111-1/+1
| * Track max session run time per Domain and return it as part of DomainInfo.Geir Storli2017-12-045-12/+46
| * Split parent + container-dependency-versions from root pom.gjoranv2017-12-011-0/+1
| * Revert "Gjoranv/split parent2"gjoranv2017-11-301-1/+0
| * Split parent + container-dependency-versions from root pom.gjoranv2017-11-301-0/+1
| * Revert "Gjoranv/split parent"gjoranv2017-11-291-1/+0
| * Merge pull request #4304 from vespa-engine/gjoranv/split-parentgjoranv2017-11-291-0/+1
| |\
| | * Split parent + container-dependency-versions from root pom.gjoranv2017-11-291-0/+1
| * | update external code (getType->type)Håvard Pettersen2017-11-283-5/+5
| |/
* | Check agreement between TF and Vespa executionJon Bratseth2017-12-1324-1441/+1767
* | Import and return constant tensorsJon Bratseth2017-12-025-47/+100
* | Import tensor constantsJon Bratseth2017-12-014-36/+140
* | Load using SavedModelBundleJon Bratseth2017-12-013-18/+19
* | Don't write to System.out/errJon Bratseth2017-11-302-16/+42
* | Remove dead codeJon Bratseth2017-11-301-17/+0
* | RefactorJon Bratseth2017-11-304-136/+161
* | More functions. Cleanup.Jon Bratseth2017-11-292-39/+53
* | Complete prototype TensorFlow mappingJon Bratseth2017-11-284-39/+212
* | Merge with masterJon Bratseth2017-11-2831-77/+495
|\|
| * Replace min/max on tensors with reduce in config modelLester Solbakken2017-11-211-1/+1
| * Enable Java tensor conformance testLester Solbakken2017-11-163-36/+50