Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Make DirectTensorAttribute::set_tensor member function private. | Tor Egge | 2023-03-02 | 1 | -1/+1 |
| | |||||
* | Test closest feature setup failures. | Tor Egge | 2023-02-27 | 1 | -1/+60 |
| | |||||
* | Rename f2 to f when appropriate. | Tor Egge | 2023-02-27 | 1 | -15/+15 |
| | | | | | Only allow mixed tensor with 1 sparse and 1 dense dimension. Return empty tensor if any distance calculator is missing. | ||||
* | Add closest feature. | Tor Egge | 2023-02-27 | 2 | -0/+159 |
| | |||||
* | Use gtest for closeness and distance features unit tests. | Tor Egge | 2023-02-27 | 2 | -53/+138 |
| | |||||
* | - Improve sanity checking of input to distance-to-path feature. | Henning Baldersheim | 2023-02-06 | 1 | -4/+5 |
| | | | | - Keep comment | ||||
* | Deinline large destructors and clean up some code based on clion hints. | Henning Baldersheim | 2023-02-01 | 1 | -0/+3 |
| | |||||
* | GC some compiler pragmas not needed. | Henning Baldersheim | 2023-01-30 | 1 | -38/+3 |
| | |||||
* | Expose SameElement query terms to ranking. | Geir Storli | 2023-01-12 | 1 | -0/+11 |
| | | | | | | A TermFieldMatchData is allocated per SameElement term, and this is used to signal matching docids in doUnpack() on the SameElement search iterator. This allows using the matches() rank feature on a field (virtual) that is searched using a SameElement term. | ||||
* | Change from typedef to using in searchlib C++ code. | Geir Storli | 2022-12-21 | 7 | -29/+29 |
| | |||||
* | Separate where you nede an actual hash, and where you want the string hashed ↵ | Henning Baldersheim | 2022-11-20 | 1 | -10/+10 |
| | | | | to a double. | ||||
* | Use AttributeBuilder in prod features unit tests. | Geir Storli | 2022-10-31 | 3 | -79/+51 |
| | |||||
* | Revert "Revert "Unify access to assets needed during rank-setup."" | Henning Baldersheim | 2022-09-07 | 1 | -11/+11 |
| | |||||
* | Revert "Unify access to assets needed during rank-setup." | Tor Egge | 2022-09-07 | 1 | -11/+11 |
| | |||||
* | Unify access to assets needed during rank-setup. | Henning Baldersheim | 2022-09-06 | 1 | -11/+11 |
| | |||||
* | Test lazy distance calculation in distance and closeness features. | Geir Storli | 2022-07-11 | 2 | -3/+60 |
| | |||||
* | Share fixture classes used by distance and closeness unit tests. | Geir Storli | 2022-07-08 | 4 | -186/+24 |
| | |||||
* | more testing (arnej) | Håvard Pettersen | 2022-06-09 | 1 | -3/+5 |
| | |||||
* | disallow bad position field | Arne H Juul | 2022-06-09 | 1 | -2/+2 |
| | |||||
* | Add bm25() and matches() and remove now() as default rank features to be ↵ | Geir Storli | 2022-06-08 | 2 | -2/+8 |
| | | | | dumped on Vespa 8. | ||||
* | Include searchcommon/attribut/config.h yourself and prepare for hiding large ↵ | Henning Baldersheim | 2022-05-25 | 8 | -8/+9 |
| | | | | | | implementation. Add setters to vespalib::GrowStrategy. | ||||
* | support default tensor values for query feature | Håvard Pettersen | 2022-05-13 | 1 | -0/+15 |
| | |||||
* | Always use MultiValueReadView for dot product feature on array attributes. | Tor Egge | 2022-04-21 | 1 | -10/+42 |
| | |||||
* | Use IMultiValueReadView in dot product feature instead of getRawValues(). | Tor Egge | 2022-04-11 | 1 | -2/+2 |
| | |||||
* | use getopt/getopt_long directly | Håvard Pettersen | 2022-04-01 | 1 | -6/+5 |
| | |||||
* | Remove inlining warnings (searchlib). | Tor Egge | 2022-02-26 | 8 | -0/+27 |
| | |||||
* | add unit tests for new features | Arne H Juul | 2022-01-31 | 2 | -1/+67 |
| | |||||
* | auto-unbox scalar results from interpreted ranking expressions | Håvard Pettersen | 2022-01-27 | 1 | -2/+6 |
| | |||||
* | added new rank feature "onnx(model_name)" | Håvard Pettersen | 2021-11-02 | 1 | -6/+23 |
| | | | | | | | | | | | Works the same as "onnxModel(model_name)", but is not treated as the exact same feature (features are currently not allowed to have multiple base names). If both variants are used at the same time, the model may be calculated twice, but the model cache will still make sure that the model itself is only loaded once. The plan is to deprecate and possibly remove the "onnxModel(model_name)" variant at some point in the future. | ||||
* | Update 2019 Oath copyrights. | gjoranv | 2021-10-27 | 2 | -2/+2 |
| | |||||
* | Don't inline OnnxFeatureTest destructor. | Tor Egge | 2021-10-27 | 1 | -0/+3 |
| | |||||
* | report onnx model evaluation as an issue | Håvard Pettersen | 2021-10-13 | 1 | -0/+14 |
| | |||||
* | Update Verizon Media copyright notices. | gjoranv | 2021-10-07 | 8 | -8/+8 |
| | |||||
* | Update 2017 copyright notices. | gjoranv | 2021-10-07 | 50 | -50/+50 |
| | |||||
* | avoid crash on run-time onnx errors | Håvard Pettersen | 2021-08-11 | 3 | -9/+12 |
| | | | | | | | | - warn about onnx model dry-run being disabled - catch and report onnx errors during ranking - zero-fill failed results to avoid re-using previous results - use explicit output size in fragile model (output became float[2] instead of float[batch] anyways) | ||||
* | dry run onnx models on setup | Håvard Pettersen | 2021-06-23 | 4 | -3/+78 |
| | |||||
* | also test tensorFromLabels with single-value integer attribute | Arne Juul | 2021-06-18 | 1 | -0/+18 |
| | |||||
* | use explicit expression name | Håvard Pettersen | 2021-06-02 | 1 | -11/+11 |
| | |||||
* | external ranking expressions | Håvard Pettersen | 2021-05-20 | 1 | -6/+23 |
| | | | | loaded from potentially compressed files | ||||
* | Revert "external ranking expressions" | Harald Musum | 2021-05-20 | 1 | -23/+6 |
| | |||||
* | external ranking expressions | Håvard Pettersen | 2021-05-18 | 1 | -6/+23 |
| | | | | loaded from potentially compressed files | ||||
* | support functional constants | Håvard Pettersen | 2021-05-11 | 1 | -1/+45 |
| | | | | | | | | | | | | be more specific about const number values (GDBT trees): is_const -> is_const_double get_const_value -> get_const_double_value Add more generic 'get_const_value' that can also be used for tensor values. Allow it to be called even in the case of parse error, in which case it does not produce a value (same as for non-const sub-expressions). | ||||
* | Use int for FastOS_UNIX_Application::GetOpt() return value. | Tor Egge | 2021-04-30 | 1 | -1/+1 |
| | |||||
* | all scalars must be double | Håvard Pettersen | 2021-03-04 | 2 | -3/+3 |
| | | | | and all operation results must be at least float | ||||
* | - Add the cost to expensive tests. | Henning Baldersheim | 2021-01-13 | 1 | -1/+1 |
| | | | | - Use less resources on the fnet_parallel_rpc test bu just using as many threads as you have cores. | ||||
* | use SimpleValue instead of EngineOrFactory | Arne Juul | 2020-12-02 | 5 | -31/+34 |
| | |||||
* | avoid using vespalib::tensor directly | Arne Juul | 2020-11-27 | 1 | -6/+14 |
| | | | | | | | * do not construct DenseTensor instances directly * instead, use TensorSpec::from_expr to make a spec * and factory.from_spec() to make a value * also, use value codec directly to encode a value | ||||
* | use factory.from_spec() instead of tensor::makeTensor<T> | Arne Juul | 2020-11-27 | 1 | -29/+29 |
| | |||||
* | cosmetic cleanups | Arne Juul | 2020-10-15 | 2 | -4/+2 |
| | |||||
* | simplify | Arne Juul | 2020-10-15 | 2 | -6/+2 |
| |