Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | TensorFlow: update eigen to latest release that has a fix to too large frame. | 2016-03-18 | |
| | | | | Change: 117506296 | ||
* | Added basic support for float16 on CPUs and older GPUs. | 2016-03-18 | |
| | | | | | Also fixed compilation issues with cuda devices that support the compute model 5.3 Change: 117493644 | ||
* | Upgraded to a newer version of Eigen that fixes a compilation error on Android | 2016-03-09 | |
| | | | | Change: 116831720 | ||
* | Upgraded eigen to make it possible to compile a binary that takes advantage of | 2016-03-09 | |
| | | | | | both avx instructions and cuda to run as fast as possible. Change: 116775924 | ||
* | Upgraded to a new version of Eigen that adds the ability to pad using values | 2016-03-08 | |
| | | | | | other than 0 and significantly speeds up a number of computations on GPUs. Change: 116607765 | ||
* | Added the ability to convert between floats and float16 on Kepler and Maxwell | 2016-03-05 | |
| | | | | | GPUs Change: 116409601 | ||
* | Improved the performance of outer reductions | 2016-03-01 | |
| | | | | Change: 116063261 | ||
* | Improved the performance of narrow reductions on CUDA | 2016-02-29 | |
| | | | | Change: 115889721 | ||
* | Upgraded Eigen to fix a compilation error triggered by xcode | 2016-02-22 | |
| | | | | Change: 115280348 | ||
* | Upgraded to the latest version of eigen, which adds a missing #include | 2016-02-22 | |
| | | | | Change: 115268843 | ||
* | Added support for half floats to eigen, which is the first step to support half | 2016-02-22 | |
| | | | | | floats in TensorFlow. The code was tested on Tegra x1. Change: 115253733 | ||
* | TensorFlow: update Eigen to non-broken GPU commit. | 2016-02-12 | |
| | | | | Change: 114585944 | ||
* | Switch the slow path in matrix_solve_ls to using ↵ | 2016-02-11 | |
| | | | | | | | Eigen::CompleteOrthogonalDecomposition (COD), which I recently contributed to Eigen in https://bitbucket.org/eigen/eigen/pull-requests/163/implement-complete-orthogonal/diff The advantage of COD over column pivoted QR is that it is able to compute the minimum-norm solution when the matrix is rank-deficient, which is usually the desired behavior and makes it consistent with the fast path. Change: 114483303 | ||
* | TensorFlow: change int -> size_t to silence warning | 2016-02-09 | |
| | | | | Change: 114243879 | ||
* | Upgraded to the latest version of Eigen that brings misc performance and | 2016-02-04 | |
| | | | | | stability improvements Change: 113791782 | ||
* | Backported several changes from the upstream version of Eigen | 2016-01-29 | |
| | | | | Change: 113371678 | ||
* | Fix OSS builds. | 2016-01-26 | |
| | | | | Change: 113114631 | ||
* | Running our linter on a lot of files. | 2016-01-24 | |
| | | | | Change: 112920860 | ||
* | Improved the performance of tensor contractions and spatial convolutions on CPU: | 2016-01-22 | |
| | | | | | | * Move some checks out of inner loops * Split the mapper in 2: a base mapper, and a sub-mapper. This reduces the number of variables that are contained in the base mapper and helps reduce register spills Change: 112809881 | ||
* | Upgrade to the latest version of eigen which improves the performance of | 2016-01-15 | |
| | | | | | matrix-vector products and provides various fixes for CUDA. Change: 112298067 | ||
* | Change: 111880705 | 2016-01-11 | |
| | |||
* | Upgraded to the latest version of Eigen, which fixes a compilation error as | 2016-01-11 | |
| | | | | | well the crash reported in https://github.com/tensorflow/tensorflow/issues/713. Change: 111874622 | ||
* | Rollback of "Pulled the latest version of Eigen, which offers much faster ↵ | 2016-01-08 | |
| | | | | | | reductions on GPU" Change: 111739152 | ||
* | Added 'logging' import to control_flow_ops which is used in the file but not ↵ | 2016-01-05 | |
| | | | | | | imported. Change: 110842260 | ||
* | Switched to the open source version of Eigen | 2015-12-16 | |
Change: 110406666 |