summaryrefslogtreecommitdiffstats
path: root/model-integration
Commit message (Collapse)AuthorAgeFilesLines
* Rename constant dimensionsJon Bratseth2019-07-2812-57/+6755
|
* Merge pull request #9996 from ↵Lester Solbakken2019-07-093-24/+24
|\ | | | | | | | | vespa-engine/havardpe/propagate-float-cell-type-and-avoid-using-it-too-much-right-now propagate float and stop using it too much
| * propagate float and stop using it too muchHåvard Pettersen2019-07-093-24/+24
| |
* | Consider all dimensionsJon Bratseth2019-07-041-12/+16
| |
* | Implement toFullStringJon Bratseth2019-07-0423-47/+88
| |
* | RefactorJon Bratseth2019-07-041-40/+49
| |
* | Try renaming under all root operations (only)Jon Bratseth2019-07-0422-25/+225
| |
* | SimplifyJon Bratseth2019-07-041-9/+6
| |
* | Loop and renameJon Bratseth2019-07-041-20/+26
| |
* | Refactor out solving without soft constraintsJon Bratseth2019-07-041-13/+12
| |
* | Cleanup commentsJon Bratseth2019-07-042-29/+29
| |
* | Keep solution possibilities internal to solverJon Bratseth2019-07-042-47/+42
| |
* | Eliminate renames reuseJon Bratseth2019-07-042-30/+28
| |
* | Extract constraint solverJon Bratseth2019-07-042-83/+117
| |
* | Inject rename to relax constraint proof of conceptJon Bratseth2019-07-0310-26/+122
| |
* | refactorJon Bratseth2019-07-031-11/+17
| |
* | Add inequality constraintsJon Bratseth2019-07-026-40/+62
| |
* | Add notEqual constraints for dimensions of the same tensor argumentJon Bratseth2019-07-023-12/+53
| |
* | Forfeit soft constraints when necessaryJon Bratseth2019-07-021-32/+45
| |
* | Forfeit soft constraints when necessaryJon Bratseth2019-07-026-41/+56
| |
* | Output conflixcting constraintsJon Bratseth2019-07-027-40/+102
| |
* | Allow extending beyond the last tensor dimensionJon Bratseth2019-07-025-13/+16
| |
* | Output intermediate graph with type info on errorJon Bratseth2019-07-018-18/+17
| |
* | Output the intermediate graphJon Bratseth2019-06-301-0/+1318
| |
* | Output the intermediate graphJon Bratseth2019-06-3011-1/+134
|/
* Merge pull request #9846 from vespa-engine/lesters/add-tf-softmax-operationLester Solbakken2019-06-2011-5/+2110
|\ | | | | Add TensorFlow softmax operation support - MERGEOK
| * Change assert message for skipped outputs in TF importLester Solbakken2019-06-204-4/+4
| |
| * Add javadoc to softmaxLester Solbakken2019-06-201-0/+5
| |
| * Remove outdated commentLester Solbakken2019-06-191-3/+1
| |
| * Add TensorFlow softmax operation supportLester Solbakken2019-06-198-2/+2104
| |
* | Revert "Merge pull request #9849 from ↵Jon Bratseth2019-06-201-2/+2
| | | | | | | | | | | | | | vespa-engine/revert-9836-bratseth/output-in-dense-tensor-form" This reverts commit 2663557e0c6d652fd082507878059a69c75e2415, reversing changes made to 46cfd5c26c5dac150efb698d726810ade4364e70.
* | Revert "Output bound dense tensors in dense string form"Jon Bratseth2019-06-191-2/+2
| |
* | Output bound dense tensors in dense string formJon Bratseth2019-06-191-2/+2
| |
* | Revert "Revert "Require constant() for large constants and fix a type ↵Jon Bratseth2019-06-132-2/+2
| | | | | | | | resolving bug""
* | Revert "Require constant() for large constants and fix a type resolving bug ↵Harald Musum2019-06-122-2/+2
| | | | | | | | | | | | (#9769)" This reverts commit cee1c3a3804d5d3c25407b3c4ac64228e9d194e3.
* | Require constant() for large constants and fix a type resolving bug (#9769)Jon Bratseth2019-06-122-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Require constant() for large constants and fix a type resolving bug * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise * Remove noise
* | Rewrite large constant references from foo to constant(foo)Jon Bratseth2019-06-061-3/+0
|/
* Expression file references in Vespa modelsJon Bratseth2019-06-064-8/+45
|
* Remove stdout debug printLester Solbakken2019-06-041-1/+0
|
* Support a few more tensorflow operationsLester Solbakken2019-06-043-1/+129
|
* Handle nested namesJon Bratseth2019-06-032-3/+5
|
* Test evaluationJon Bratseth2019-06-032-4/+26
|
* Vespa global model importJon Bratseth2019-06-027-43/+86
|
* Support small constantsJon Bratseth2019-05-313-77/+84
|
* Support native Vespa standalone modelsJon Bratseth2019-05-3116-30/+515
|
* Revert "Bratseth/emit float tensors in config"Arnstein Ressem2019-05-081-3/+3
|
* Merge pull request #9307 from vespa-engine/bratseth/emit-float-tensors-in-configLester Solbakken2019-05-081-3/+3
|\ | | | | Bratseth/emit float tensors in config
| * Expect float tensors from OnnxJon Bratseth2019-05-071-3/+3
| |
* | ONNX: use node output as name if not explicitly setLester Solbakken2019-05-074-11/+137
|/
* Use value type of ONNX tensor argumentsJon Bratseth2019-04-046-83/+75
|