Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Fixed compilation errors triggered by old versions of gcc | 2016-05-12 | |
| | |||
* | Eigen cost model part 1. This implements a basic recursive framework to ↵ | 2016-04-14 | |
| | | | | estimate the cost of evaluating tensor expressions. | ||
* | Worked around a constness related issue | 2016-03-21 | |
| | |||
* | Small cleanup | 2016-03-21 | |
| | |||
* | Decoupled the packet type definition from the definition of the tensor ops. ↵ | 2016-03-08 | |
| | | | | All the vectorization is now defined in the tensor evaluators. This will make it possible to relialably support devices with different packet types in the same compilation unit. | ||
* | Deleted the coordinate based evaluation of tensor expressions, since it's ↵ | 2016-02-22 | |
| | | | | hardly ever used and started to cause some issues with some versions of xcode. | ||
* | Fixed a constness bug | 2016-01-21 | |
| | |||
* | Record whether the underlying tensor storage can be accessed directly during ↵ | 2016-01-19 | |
| | | | | the evaluation of an expression. | ||
* | Add alignment requirement for local buffer used by the slicing op. | 2015-12-18 | |
| | |||
* | Fix return type for TensorEvaluator<TensorSlicingOp>::data | 2015-09-17 | |
| | |||
* | Many files were missing in previous changeset. | 2015-07-29 | |
| | |||
* | Use NumTraits<T>::RequireInitialization instead of ↵ | 2015-07-07 | |
| | | | | internal::is_arithmetic<T>::value to check whether it's possible to bypass the type constructor in the tensor code. | ||
* | Misc small fixes to the tensor slicing code. | 2015-07-06 | |
| | |||
* | Fixed a couple of mistakes in the previous commit. | 2015-07-01 | |
| | |||
* | Enabled the vectorized evaluation of several tensor expressions that was ↵ | 2015-07-01 | |
| | | | | previously disabled by mistake | ||
* | Avoid using the cuda memcpy for small tensor slices since the memcpy kernel ↵ | 2015-05-19 | |
| | | | | is very expensive to launch | ||
* | Sped up the assignment of a tensor to a tensor slice, as well as the ↵ | 2015-04-20 | |
| | | | | assigment of a constant slice to a tensor | ||
* | Fixed another batch of compilation warnings | 2015-02-28 | |
| | |||
* | Fixed several compilation warnings reported by clang | 2015-02-25 | |
| | |||
* | Silenced a few compilation warnings generated by nvcc | 2015-02-10 | |
| | |||
* | Silenced more compilation warnings | 2015-02-10 | |
| | |||
* | gcc doesn't consider that | 2015-01-16 | |
| | | | | | | | | | | | template<typename OtherDerived> TensorStridingOp& operator = (const OtherDerived& other) provides a valid assignment operator for the striding operation, and therefore refuses to compile code like: result.stride(foo) = source.stride(bar); Added the explicit TensorStridingOp& operator = (const TensorStridingOp& other) as a workaround to get the code to compile, and did the same in all the operations that can be used as lvalues. | ||
* | Improved support for RowMajor tensors | 2015-01-14 | |
| | | | | Misc fixes and API cleanups. | ||
* | fixed some potential alignment issues. | 2014-10-30 | |
| | |||
* | Added access to the unerlying raw data of a tnsor slice/chip whenever possible | 2014-10-17 | |
| | |||
* | Silenced a few compilation warnings | 2014-10-16 | |
| | | | | Generalized a TensorMap constructor | ||
* | Added tests for tensors of const values and tensors of stringswwq:: | 2014-10-01 | |
| | |||
* | Misc fixes. | 2014-09-05 | |
| | |||
* | Optimized and cleaned up the tensor morphing code | 2014-08-23 | |
| | |||
* | Added support for fast integer divisions by a constant | 2014-08-14 | |
| | | | | Sped up tensor slicing by a factor of 3 by using these fast integer divisions. | ||
* | Support for in place evaluation of expressions containing slicing and ↵ | 2014-08-13 | |
| | | | | reshaping operations | ||
* | Vectorized the evaluation of expressions involving tensor slices. | 2014-07-10 | |
| | |||
* | Improved the speed of slicing operations. | 2014-07-09 | |
| | |||
* | Added support for tensor slicing | 2014-07-07 | |
| | |||
* | Reworked the expression evaluation mechanism in order to make it possible to ↵ | 2014-06-13 | |
| | | | | | | | | efficiently compute convolutions and contractions in the future: * The scheduling of computation is moved out the the assignment code and into a new TensorExecutor class * The assignment itself is now a regular node on the expression tree * The expression evaluators start by recursively evaluating all their subexpressions if needed | ||
* | TensorEval are now typed on the device: this will make it possible to use ↵ | 2014-06-10 | |
| | | | | | | partial template specialization to optimize the strategy of each evaluator for each device type. Started work on partial evaluations. | ||
* | Improved support for rvalues in tensor expressions. | 2014-06-09 | |
| | |||
* | Added support for convolution and reshaping of tensors. | 2014-06-06 | |