Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Serialize scalar functions with context | Jon Bratseth | 2019-11-08 | 2 | -18/+39 |
| | |||||
* | Always serialize dynamic tensors in verbose form | Jon Bratseth | 2019-11-08 | 3 | -4/+30 |
| | |||||
* | Literal tensors with value expressions | Jon Bratseth | 2019-11-05 | 2 | -2/+146 |
| | |||||
* | Link as link | Jon Bratseth | 2019-10-08 | 1 | -1/+2 |
| | |||||
* | Link as link | Jon Bratseth | 2019-10-08 | 1 | -1/+1 |
| | |||||
* | Single sparse dimension short form | Jon Bratseth | 2019-10-08 | 1 | -11/+33 |
| | |||||
* | Simplify | Jon Bratseth | 2019-10-08 | 1 | -7/+5 |
| | |||||
* | Support mixed tensor short form JSON | Jon Bratseth | 2019-10-08 | 1 | -35/+51 |
| | |||||
* | Support mixed tensor short form JSON | Jon Bratseth | 2019-10-08 | 3 | -8/+47 |
| | |||||
* | Document the standard value order one place | Jon Bratseth | 2019-10-01 | 4 | -36/+34 |
| | |||||
* | use same rules for cell value type resolving as C++ | Arne Juul | 2019-08-20 | 4 | -9/+18 |
| | | | | | * pick cell value type from tensors with dimensions only * in Concat, use the expected combined cell value type for unit tensor | ||||
* | Merge | Jon Bratseth | 2019-08-13 | 2 | -18/+26 |
| | |||||
* | propagate float and stop using it too much | Håvard Pettersen | 2019-07-09 | 1 | -1/+6 |
| | |||||
* | Convenience method to find smallest/largest cells | Jon Bratseth | 2019-07-03 | 2 | -1/+55 |
| | |||||
* | Use value accessor in convenience accessors | Jon Bratseth | 2019-07-02 | 1 | -2/+2 |
| | |||||
* | Keep writing floats as doubles in toString for now | Jon Bratseth | 2019-06-20 | 1 | -1/+1 |
| | |||||
* | Revert "Merge pull request #9849 from ↵ | Jon Bratseth | 2019-06-20 | 1 | -1/+51 |
| | | | | | | | vespa-engine/revert-9836-bratseth/output-in-dense-tensor-form" This reverts commit 2663557e0c6d652fd082507878059a69c75e2415, reversing changes made to 46cfd5c26c5dac150efb698d726810ade4364e70. | ||||
* | Read dense tensor form in documents | Jon Bratseth | 2019-06-18 | 2 | -0/+6 |
| | |||||
* | Deserialize dense form | Jon Bratseth | 2019-06-18 | 1 | -6/+32 |
| | |||||
* | Decode tensor summary features | Jon Bratseth | 2019-06-15 | 1 | -4/+16 |
| | |||||
* | Parse dense tensors in the rightmost adjacent order | Jon Bratseth | 2019-06-11 | 2 | -49/+36 |
| | |||||
* | Dense string form | Jon Bratseth | 2019-06-07 | 5 | -46/+140 |
| | |||||
* | Expect IllegalArgumentException and be consistent | Jon Bratseth | 2019-06-04 | 2 | -11/+21 |
| | |||||
* | Adjust to changes exception type | Jon Bratseth | 2019-06-04 | 1 | -1/+1 |
| | |||||
* | Validate more | Jon Bratseth | 2019-06-04 | 5 | -12/+24 |
| | |||||
* | Merge pull request #9641 from vespa-engine/bratseth/ranking-expression-models | Jon Bratseth | 2019-06-02 | 2 | -3/+27 |
|\ | | | | | Bratseth/ranking expression models | ||||
| * | Vespa global model import | Jon Bratseth | 2019-06-02 | 1 | -3/+20 |
| | | |||||
| * | Support native Vespa standalone models | Jon Bratseth | 2019-05-31 | 1 | -0/+7 |
| | | |||||
* | | Update ABI spec and add javadocW | Jon Bratseth | 2019-06-01 | 3 | -6/+42 |
| | | |||||
* | | Allow passing your own vector without copy to the IndexedTensor. | Henning Baldersheim | 2019-06-01 | 3 | -12/+79 |
|/ | |||||
* | Revert "Bratseth/emit float tensors in config" | Arnstein Ressem | 2019-05-08 | 2 | -26/+14 |
| | |||||
* | Emit float tensor types in config when specified | Jon Bratseth | 2019-05-07 | 2 | -14/+26 |
| | |||||
* | Add a test and correct condition | Jon Bratseth | 2019-04-29 | 1 | -1/+1 |
| | |||||
* | Add float accessor | Jon Bratseth | 2019-04-26 | 1 | -0/+10 |
| | |||||
* | Encode directly as float | Jon Bratseth | 2019-04-26 | 4 | -28/+25 |
| | |||||
* | Decode directly as float | Jon Bratseth | 2019-04-26 | 1 | -8/+20 |
| | |||||
* | Build tensors purely with floats | Jon Bratseth | 2019-04-26 | 6 | -17/+100 |
| | |||||
* | Make float builder when appropriate | Jon Bratseth | 2019-04-26 | 3 | -7/+59 |
| | |||||
* | Move bound builder double array into double subclass | Jon Bratseth | 2019-04-26 | 3 | -60/+128 |
| | |||||
* | Split values into IndexedDoubleTensor subclass | Jon Bratseth | 2019-04-25 | 3 | -34/+64 |
| | |||||
* | Add tensor value serialization for mapped and mixed tensors | Lester Solbakken | 2019-04-11 | 4 | -112/+174 |
| | |||||
* | Use bit number 3 mean 'with cell type' to carry forward the current schema ↵ | Henning Baldersheim | 2019-04-05 | 1 | -3/+5 |
| | | | | | | where the bits hav meaning. Update expected result. | ||||
* | Infer value type | Jon Bratseth | 2019-04-03 | 2 | -8/+8 |
| | |||||
* | Consider value type in equals | Jon Bratseth | 2019-04-03 | 1 | -4/+8 |
| | |||||
* | Add tensor value type | Jon Bratseth | 2019-04-03 | 10 | -30/+63 |
| | |||||
* | Parse tensor value type | Jon Bratseth | 2019-04-02 | 4 | -43/+80 |
| | |||||
* | Test and ensure that serialization is symmetric. | Henning Baldersheim | 2019-04-02 | 1 | -1/+1 |
| | |||||
* | Verify serialization of various cell value types. | Henning Baldersheim | 2019-04-01 | 1 | -1/+7 |
| | |||||
* | Add support for different values during decoding too. | Henning Baldersheim | 2019-04-01 | 2 | -11/+59 |
| | |||||
* | Initial skeleton for type information in serialization format. | Henning Baldersheim | 2019-04-01 | 3 | -3/+62 |
| |