Commit message (Expand) | Author | Age | |
---|---|---|---|
* | Fix python 3 GPU test failures | 2018-09-10 | |
* | Add needed specifications for Split on GPU. | 2018-09-07 | |
* | Optimized non-aligned case of split and split_v on the first input dimension. | 2018-02-27 | |
* | add bfloat16 support to some GPU ops: concat, constant, fill, pack, reshape, | 2017-12-15 | |
* | More WhereOp/TopK GPU bugfixes: use the direct cuda stream for CUB GPU kernel. | 2017-07-07 | |
* | Add Split and SplitV GPU support for complex64/complex128. | 2017-04-10 | |
* | Add a new op split_v that can handle variable size splits. | 2016-11-10 | |
* | Update copyright for 3p/tf/core. | 2016-06-02 | |
* | In SplitOp GPU, avoid memcpying the pointers to the device if the number of | 2016-05-10 | |
* | Give tensorflow/core/kernels/ its own BUILD file. | 2016-02-24 |