| Commit message (Expand) | Author | Age |
* | [ROCm] Interface changes for pooling APIs in StreamExecutor | Wen-Heng (Jack) Chung | 2018-07-11 |
* | Fix Windows GPU Build | A. Unique TensorFlower | 2018-06-28 |
* | Reflow comments; NFC | Sanjoy Das | 2018-06-15 |
* | Merge changes from github. | Yifei Feng | 2018-05-24 |
* | Small polishing changes in stream executor, no functional changes. | A. Unique TensorFlower | 2018-05-15 |
* | Use parenthesis based construction instead of brace initialization | Smit Hinsu | 2018-05-09 |
* | Add variants of DoBlasGemmWithAlgorithm with alpha being on device. | A. Unique TensorFlower | 2018-04-24 |
* | [StreamExecutor] Rename ::perftools::gputools -> ::stream_executor, part 1. | Justin Lebar | 2018-04-17 |
* | Support RNN profiling in StreamExecutor for CUDA GPUs. | James Qin | 2018-04-06 |
* | [StreamExecutor] Remove ThenDoHostCallbackForTest -- it's identical to ThenDo... | Justin Lebar | 2018-03-09 |
* | StreamExecutor support for float64 convolutions and backprop. | Brian Patton | 2018-03-06 |
* | [StreamExecutor] Change "variance" to "inv_var" in BatchNormalizationBackward. | Justin Lebar | 2017-12-18 |
* | Remove Stream::BlockHostUntilDoneWithStatus; all callers use BlockHostUntilDone. | A. Unique TensorFlower | 2017-12-15 |
* | Stream::BlockHostUntilDone now returns Status rather than bool. | A. Unique TensorFlower | 2017-12-13 |
* | Add BlockHostUntilDoneWithStatus, which returns Status rather than bool. | A. Unique TensorFlower | 2017-12-06 |
* | Support Cudnn RNN Fp16 | James Qin | 2017-11-03 |
* | Merge changes from github. | Frank Chen | 2017-10-06 |
* | Add float16 support to tf.nn.fused_batch_norm on the GPU. | Reed Wanderman-Milne | 2017-09-27 |
* | Merge changes from github. | Shanqing Cai | 2017-09-25 |
* | Add int8 version of fused_conv2d_bias_activation operator for the forward phase, | A. Unique TensorFlower | 2017-09-06 |
* | Automated g4 rollback of changelist 166276461 | A. Unique TensorFlower | 2017-08-24 |
* | Add int8 version of fused_conv2d_bias_activation operator for the forward phase, | A. Unique TensorFlower | 2017-08-23 |
* | Make tensorflow::mutex implement a shared (reader/writer) lock, using | A. Unique TensorFlower | 2017-08-17 |
* | Let GetBlasGemmAlgorithms() always return true. | Yangzihao Wang | 2017-07-21 |
* | Automated g4 rollback of changelist 162423171 | A. Unique TensorFlower | 2017-07-18 |
* | Add autotuning code for matmul operator. | Yangzihao Wang | 2017-07-18 |
* | Support float64 CuDNN RNN | James Qin | 2017-07-18 |
* | Add support for int8 x int8 -> int32 matrix multiplication via cublasGemmEx t... | A. Unique TensorFlower | 2017-07-06 |
* | [SE] Support alpha scale in cudnnTransformTensor | A. Unique TensorFlower | 2017-06-20 |
* | TransformTensor supports NCHW_VECT_C layout and int8 data type. | Jingyue Wu | 2017-06-12 |
* | Pass int parameter by value, not by const reference | A. Unique TensorFlower | 2017-06-06 |
* | [SE] Add cudnnTransformTensor to StreamExecutor. | Jingyue Wu | 2017-06-05 |
* | Add functional support for cudnnConvolutionBiasActivationForward(). | Yangzihao Wang | 2017-06-01 |
* | Merge changes from github. | A. Unique TensorFlower | 2017-04-04 |
* | [XLA] [StreamExecutor] Tune GEMMs when possible. | Justin Lebar | 2017-03-02 |
* | [StreamExecutor] Minor comment cleanups. | Justin Lebar | 2017-03-02 |
* | [XLA:GPU] Cache GPU substreams across executions | A. Unique TensorFlower | 2017-03-02 |
* | Add options argument for DNN activation | A. Unique TensorFlower | 2017-01-24 |
* | Add convolve quantized ops to StreamExecutor API | A. Unique TensorFlower | 2017-01-19 |
* | Add several operations to the StreamExecutor API | A. Unique TensorFlower | 2017-01-17 |
* | Add the interface in steam executor to call cuDNN batch normalization functions. | Yao Zhang | 2016-09-15 |
* | Add stream-executor changes to enable Cudnn fused LSTM/RNN support. | Xiaoqiang Zheng | 2016-08-19 |
* | Roll-forward of "Local Response Normalization GPU support via Stream Executor." | RJ Ryan | 2016-07-13 |
* | Automated rollback of change 127123966 | Vijay Vasudevan | 2016-07-11 |
* | Local Response Normalization GPU support via Stream Executor. | RJ Ryan | 2016-07-11 |
* | Improve convolution autotune process. The max batch size VGG model can handle | Xiaoqiang Zheng | 2016-06-21 |
* | Enable fp16 for most of the pooling ops (MaxPool, AvgPool, associated | Benoit Steiner | 2016-06-06 |
* | Merge changes from github. | Martin Wicke | 2016-06-06 |
* | Enable fp16 for most of the pooling ops (MaxPool, AvgPool, associated | A. Unique TensorFlower | 2016-06-03 |
* | Enable fp16 for most of the pooling ops (MaxPool, AvgPool, associated | Benoit Steiner | 2016-06-03 |