Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Revert "Copyright header" | Jon Bratseth | 2017-06-13 | 208 | -208/+208 |
| | |||||
* | Copyright header | Jon Bratseth | 2017-06-13 | 208 | -208/+208 |
| | |||||
* | auto-detect appropriate compiled function parameter lazyness | Haavard | 2017-06-07 | 6 | -0/+89 |
| | | | | ... but still allow config override | ||||
* | Add support for conditionally running available shutdown. | Henning Baldersheim | 2017-06-07 | 1 | -0/+5 |
| | |||||
* | Include stringfmt where it was needed. | Henning Baldersheim | 2017-06-06 | 1 | -0/+1 |
| | |||||
* | cleanup a global mutex | Arne H Juul | 2017-06-02 | 1 | -0/+1 |
| | |||||
* | Reduce fastos.h dependency. | Henning Baldersheim | 2017-06-01 | 48 | -70/+43 |
| | |||||
* | call llvm_shutdown | Haavard | 2017-05-29 | 1 | -0/+3 |
| | |||||
* | clone for wrapped tensors | Haavard | 2017-05-24 | 4 | -10/+10 |
| | |||||
* | support tensor mapping for wrapped simple tensors | Haavard | 2017-05-23 | 6 | -265/+367 |
| | |||||
* | wrapped simple tensor | Haavard | 2017-05-18 | 14 | -142/+312 |
| | | | | | | | | | | | | fall back to wrapping SimpleTensors for cases where the default tensor implementation does not support the data model (mixed dimensions). remove special handling of mixed cases in tensor conformance tests. require that the default tensor implementation passes all conformance tests. ironed out a small mismatch in the tensor binary format tag. | ||||
* | fixes based on review feedback | Haavard | 2017-05-11 | 1 | -7/+7 |
| | |||||
* | test tensor binary format | Haavard | 2017-05-10 | 12 | -7/+862 |
| | |||||
* | Merge pull request #2418 from yahoo/havardpe/encode-decode-simple-tensors | Geir Storli | 2017-05-09 | 8 | -134/+342 |
|\ | | | | | binary encode/decode as part of tensor engine API | ||||
| * | binary encode/decode as part of tensor engine API | Haavard | 2017-05-08 | 8 | -134/+342 |
| | | |||||
* | | Deinline as compiler makes the best choices. | Henning Baldersheim | 2017-05-09 | 4 | -0/+7 |
| | | |||||
* | | Fix remaining inline warnings for compilation with -Og | Tor Brede Vekterli | 2017-05-05 | 2 | -0/+4 |
| | | |||||
* | | gcc 7 does not provide <functional> by default. | Henning Baldersheim | 2017-05-05 | 1 | -3/+4 |
| | | |||||
* | | gcc 7 does not provide <functional> by default. | Henning Baldersheim | 2017-05-05 | 1 | -4/+5 |
|/ | |||||
* | Merge pull request #2340 from yahoo/havardpe/mixed-tensor-binary-format-spec | Geir Storli | 2017-05-03 | 1 | -0/+42 |
|\ | | | | | added description of new mixed serialization format | ||||
| * | minor fixup | Haavard | 2017-05-03 | 1 | -1/+1 |
| | | |||||
| * | account for sparse tensors without dimensions | Haavard | 2017-05-03 | 1 | -6/+11 |
| | | |||||
| * | added description of new mixed serialization format | Haavard | 2017-05-02 | 1 | -0/+37 |
| | | | | | | | | ... and how it relates to old formats (sparse/dense) | ||||
* | | Fix warnings hidden earlier due to including application headers as system ↵ | Henning Baldersheim | 2017-05-02 | 11 | -17/+40 |
|/ | | | | includes | ||||
* | Revert "Balder/enforce override 2" | Arne H Juul | 2017-04-24 | 1 | -4/+4 |
| | |||||
* | Use override | Henning Baldersheim | 2017-04-23 | 1 | -4/+4 |
| | |||||
* | add override in eval module | Arne H Juul | 2017-04-10 | 1 | -2/+3 |
| | |||||
* | Override | Henning Baldersheim | 2017-04-04 | 11 | -355/+350 |
| | |||||
* | change code to work with LLVM 3.9.1 | Haavard | 2017-03-28 | 7 | -73/+96 |
| | |||||
* | Move non-inlineable eval test functions into cpp file | Tor Brede Vekterli | 2017-03-13 | 2 | -37/+51 |
| | |||||
* | Create explicit, non implicitly inlined function definitions | Tor Brede Vekterli | 2017-03-13 | 2 | -0/+3 |
| | |||||
* | Deinline destructors/constructors | Henning Baldersheim | 2017-03-08 | 3 | -11/+23 |
| | |||||
* | deinline large destructors | Henning Baldersheim | 2017-03-08 | 2 | -7/+13 |
| | |||||
* | Specify destructors explicit to control inlining. | Henning Baldersheim | 2017-03-08 | 16 | -93/+170 |
| | | | | | Conflicts: build_settings.cmake | ||||
* | more specific initial value | Haavard | 2017-03-02 | 1 | -1/+1 |
| | |||||
* | analyze parameter usage | Haavard | 2017-03-02 | 6 | -1/+210 |
| | |||||
* | interpreted functions now always take lazy parameters | Haavard | 2017-02-27 | 8 | -74/+102 |
| | |||||
* | also log lazy time/factor in explore benchmark | Haavard | 2017-02-21 | 1 | -12/+28 |
| | |||||
* | support gbdt forest optimizations with lazy parameters | Haavard | 2017-02-21 | 7 | -61/+196 |
| | |||||
* | improve and extend micro-benchmark | Haavard | 2017-02-20 | 5 | -85/+156 |
| | |||||
* | added support for lazy parameters to compiled functions | Haavard | 2017-02-17 | 9 | -22/+54 |
| | |||||
* | convert result from map/apply to DoubleValue if type is double | Haavard | 2017-02-16 | 1 | -1/+8 |
| | |||||
* | let tensor::Tensor::getType return const reference | Haavard | 2017-02-16 | 5 | -5/+5 |
| | | | | | ... to make it cheaper to check tensor types in implementation-specific code. | ||||
* | use simple tensor engine as (expensive) fallback | Haavard | 2017-02-16 | 2 | -21/+39 |
| | | | | ... for new immediate API in default tensor engine | ||||
* | extend tensor conformance test with new syntax | Haavard | 2017-02-10 | 1 | -42/+78 |
| | |||||
* | add more reduce cases and drop tensor-only eval in double context | Haavard | 2017-02-10 | 1 | -5/+7 |
| | |||||
* | enable converting simple tensors to errors | Haavard | 2017-02-10 | 1 | -17/+11 |
| | |||||
* | allow simple tensors to represent error values | Haavard | 2017-02-10 | 2 | -0/+20 |
| | |||||
* | wire in immediate evaluation of new syntax | Haavard | 2017-02-10 | 9 | -73/+109 |
| | |||||
* | add support for loading lz4 compressed json tensors | Haavard | 2017-02-09 | 5 | -10/+39 |
| |