summaryrefslogtreecommitdiffstats
path: root/eval/src/tests/instruction/generic_concat
Commit message (Collapse)AuthorAgeFilesLines
* use CellTypeUtils::list_types to loop over possible cell types in testsArne Juul2021-03-031-6/+4
|
* use SmallVector in GenericConcatArne Juul2021-03-011-6/+6
|
* adjust param repo add options and remove seq_biasHåvard Pettersen2021-02-051-4/+4
| | | | GenSpec is now implicitly convertible to TensorSpec
* Merge pull request #16347 from vespa-engine/arnej/genspec-review-followupArne H Juul2021-02-031-6/+10
|\ | | | | followup on GenSpec after review
| * followup on GenSpec after reviewArne Juul2021-02-021-6/+10
| | | | | | | | | | | | * avoid changing layouts in-place, call cpy() first * do cells_float() vs cells_double() more equally * minor cosmetic fixes
* | Use noexcept specifier for simple lambdas.Tor Egge2021-02-021-1/+1
|/
* copy gen spec to avoid sticky changesHåvard Pettersen2021-02-011-4/+4
|
* use GenSpec in generic_concat_testArne Juul2021-02-011-54/+53
|
* forward index for concat of mixed tensor with dense tensorArne Juul2021-01-151-0/+11
|
* remove immediate APIArne Juul2020-12-031-19/+0
|
* remove comparison with SimpleTensor resultsArne Juul2020-11-301-21/+0
|
* move reference operations to a common placeArne Juul2020-11-251-60/+4
| | | | | | | * lift reference implementations from unit tests * write a reference TensorCreate implementation * use these from unit tests * some minor cosmetic changes with structured bindings
* add unit tests with FastValueBuilderFactory alsoArne Juul2020-10-221-4/+14
|
* test immediate formsArne Juul2020-10-191-0/+20
|
* allow interpreted function to use new generic operationsHåvard Pettersen2020-10-121-1/+1
| | | | update conformance test to also test with SimpleValue and FastValue
* Merge pull request #14757 from vespa-engine/add-tensor-partial-update-2Arne H Juul2020-10-091-2/+2
|\ | | | | Add tensor partial update 2
| * simplify TensorSpec::add()Arne Juul2020-10-071-2/+2
| | | | | | | | | | * the old multiple-add semantics weren't used for anything, so make it more predictable and constrained.
* | handle concat(vector of float, double)Arne Juul2020-10-081-0/+3
| |
* | do per-parameter resolving in a method and only keep the results we needArne Juul2020-10-081-4/+1
| |
* | add layouts to trigger now-fixed bugArne Juul2020-10-081-27/+35
| |
* | check entire dense concat planArne Juul2020-10-081-0/+14
| |
* | add simple unit test for dense concat planArne Juul2020-10-081-0/+12
|/
* use dimension_index instead of explicit loopArne Juul2020-10-061-8/+8
|
* also merge similar casesArne Juul2020-10-051-0/+2
|
* add generic_concat, with testArne Juul2020-10-051-3/+24
|
* add reference implementation of concat, with testArne Juul2020-10-052-0/+132