Commit message (Collapse) | Author | Age | ||
---|---|---|---|---|
... | ||||
* | Fixed a compilation warning generated by clang | 2015-07-06 | ||
| | ||||
* | Only attempt to use the texture path on GPUs when it's supported by CUDA | 2015-07-06 | ||
| | ||||
* | Misc small fixes to the tensor slicing code. | 2015-07-06 | ||
| | ||||
* | Added a default value for the cuda stream in the GpuDevice constructor | 2015-07-01 | ||
| | ||||
* | Misc improvements and optimizations | 2015-07-01 | ||
| | ||||
* | Improved a previous fix | 2015-07-01 | ||
| | ||||
* | 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 | |||
* | Silenced a compilation warning | 2015-06-30 | ||
| | ||||
* | Added support for user defined custom tensor op. | 2015-06-30 | ||
| | ||||
* | Added support for 3D patch extraction | 2015-06-30 | ||
| | ||||
* | Made ThreadPoolDevice inherit from a new pure abstract ThreadPoolInterface ↵ | 2015-06-30 | ||
| | | | | class: this enables users to leverage their existing threadpool when using eigen tensors. | |||
* | Turned Eigen::array::size into a function to make the code compatible with ↵ | 2015-06-30 | ||
| | | | | std::array | |||
* | Added a test for multithreaded full reductions | 2015-06-30 | ||
| | ||||
* | Fixed a few compilation warnings | 2015-06-30 | ||
| | ||||
* | Silenced a number of compilation warnings | 2015-06-29 | ||
| | ||||
* | Added a test for full reductions on GPU | 2015-06-29 | ||
| | ||||
* | Improved performance of full reduction by 2 order of magnitude on CPU and 3 ↵ | 2015-06-29 | ||
| | | | | orders of magnitude on GPU | |||
* | Improved support for fixed size tensors | 2015-06-29 | ||
| | ||||
* | Express the full reduction operations (such as sum, max, min) using ↵ | 2015-06-29 | ||
| | | | | TensorDimensionList | |||
* | Added support for tanh function to the tensor code | 2015-06-29 | ||
| | ||||
* | Moved some utilities to TensorMeta.h to make it easier to reuse them accross ↵ | 2015-06-29 | ||
| | | | | | | several tensor operations. Created the TensorDimensionList class to encode the list of all the dimensions of a tensor of rank n. This could be done using TensorIndexList, however TensorIndexList require cxx11 which isn't yet supported as widely as we'd like. | |||
* | Merged in vanhoucke/eigen_vanhoucke (pull request PR-118) | 2015-06-20 | ||
|\ | | | | | | | Fix two small undefined behaviors caught by static analysis. | |||
* | | Get rid of class internal::nested<> (still have to updated Tensor module) | 2015-06-19 | ||
| | | ||||
| * | Fix undefined behavior. | 2015-06-19 | ||
|/ | ||||
* | Fixed a compilation warning | 2015-06-17 | ||
| | ||||
* | Fixed merge conflict | 2015-06-16 | ||
| | ||||
* | Pulled latest updates from trunk | 2015-06-16 | ||
|\ | ||||
* | | Fixed compilation warnings triggered by clang | 2015-06-16 | ||
| | | ||||
| * | Fix compilation in TensorImagePatch | 2015-06-16 | ||
| | | ||||
| * | Fix shadow warnings in Tensor module | 2015-06-16 | ||
| | | ||||
| * | Fix compilation of alignedvector3 unit test | 2015-06-16 | ||
|/ | ||||
* | code simplification | 2015-06-09 | ||
| | ||||
* | formatting | 2015-06-09 | ||
| | ||||
* | remove redundant dynamic allocations in GMRES | 2015-06-09 | ||
| | ||||
* | fix some legitimate shadow warnings | 2015-06-09 | ||
| | ||||
* | Do not abort if the folder cannot be openned! | 2015-06-05 | ||
| | ||||
* | Improve loading of symmetric sparse matrices in MatrixMarketIterator | 2015-06-05 | ||
| | ||||
* | Fixed a compilation error triggered by nvcc 7 | 2015-05-28 | ||
| | ||||
* | Worked around some constexpr related bugs in nvcc 7 | 2015-05-28 | ||
| | ||||
* | Added missing include files | 2015-05-28 | ||
| | ||||
* | Fixed potential compilation error | 2015-05-26 | ||
| | ||||
* | Added a few more missing EIGEN_DEVICE_FUNC statements | 2015-05-26 | ||
| | ||||
* | Added a few missing EIGEN_DEVICE_FUNC statements | 2015-05-26 | ||
| | ||||
* | Moved away from std::async and std::future as the underlying mechnism for ↵ | 2015-05-20 | ||
| | | | | | | the thread pool device. On several platforms, the functions passed to std::async are not scheduled in the order in which they are given to std::async, which leads to massive performance issues in the contraction code. Instead we now have a custom thread pool that ensures that the functions are picked up by the threads in the pool in the order in which they are enqueued in the pool. | |||
* | Fixed compilation error triggered by gcc 4.7 | 2015-05-20 | ||
| | ||||
* | Avoid using the cuda memcpy for small tensor slices since the memcpy kernel ↵ | 2015-05-19 | ||
| | | | | is very expensive to launch | |||
* | Added new version of the TensorIntDiv class optimized for 32 bit signed ↵ | 2015-05-19 | ||
| | | | | integers. It saves 1 register on CPU and 2 on GPU. | |||
* | Fix trivial warnings in LevenbergMarquardt module and test | 2015-04-24 | ||
| | ||||
* | Silenced a few compilation warnings | 2015-04-22 | ||
| |