| Commit message (Expand) | Author | Age |
* | Fix cub include path so that TensorFlow compiles when used as a bazel depende... | A. Unique TensorFlower | 2018-09-20 |
* | Avoid ambiguous 'detail' namespace. | A. Unique TensorFlower | 2018-07-18 |
* | Support for pseudo half precision in depthwise convolution kernels. | A. Unique TensorFlower | 2018-07-17 |
* | Add NCHW_VECT_W tensor format. | A. Unique TensorFlower | 2018-05-17 |
* | Use tensorflow::se instead of perftools::gputools for StreamExecutor. | Justin Lebar | 2018-04-23 |
* | Merge changes from github. | Yifei Feng | 2018-02-22 |
* | Fix pontential issue with number of blocks launched for depthwise kernels: th... | A. Unique TensorFlower | 2018-02-16 |
* | Mechanical variable renaming to improve consistency. No other changes. | A. Unique TensorFlower | 2018-02-13 |
* | Getting rid of unnecessary GPUDevice typedef. | A. Unique TensorFlower | 2018-02-10 |
* | Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/core/kernels. | A. Unique TensorFlower | 2018-01-26 |
* | Roll CL 179861781 forward with fix: Wrappers for CUDA 9 warp-synchronous intr... | A. Unique TensorFlower | 2018-01-26 |
* | Automated g4 rollback of changelist 179782067 | A. Unique TensorFlower | 2017-12-21 |
* | Roll CL 177989542 forward with fix: Wrappers for CUDA 9 warp-synchronous intr... | A. Unique TensorFlower | 2017-12-21 |
* | Automated g4 rollback of changelist 177799252 | A. Unique TensorFlower | 2017-12-05 |
* | Wrappers for CUDA 9 warp-synchronous intrinsics. | A. Unique TensorFlower | 2017-12-04 |
* | Merge changes from github. | Yifei Feng | 2017-11-22 |
* | Automated g4 rollback of changelist 176615107 | Yifei Feng | 2017-11-22 |
* | Merge changes from github. | Yifei Feng | 2017-11-21 |
* | Merge changes from github. | Patrick Nguyen | 2017-09-15 |
* | Exposing launchpad for conv2d backprop, and unify launchpads for conv2d and d... | A. Unique TensorFlower | 2017-09-04 |
* | In the CUDA path of depthwise_conv2d, merge 'Forward' and 'BackpropInput' var... | A. Unique TensorFlower | 2017-06-19 |
* | Automated g4 rollback of changelist 158990733 | A. Unique TensorFlower | 2017-06-15 |
* | Automated g4 rollback of changelist 158765635 | Vijay Vasudevan | 2017-06-14 |
* | In the CUDA path of depthwise_conv2d, extend the fastpath to support tensor w... | A. Unique TensorFlower | 2017-06-12 |
* | In the CUDA path of depthwise_conv2d, optimize backward filter convolution fo... | A. Unique TensorFlower | 2017-06-07 |
* | In the CUDA path of depthwise_conv2d, add a fast NCHW backward filter convolu... | A. Unique TensorFlower | 2017-06-06 |
* | In the CUDA path of depthwise_conv2d, add a fast NCHW backward input convolut... | A. Unique TensorFlower | 2017-06-05 |
* | In the CUDA path of depthwise_conv2d, add a fast NCHW forward convolution for... | A. Unique TensorFlower | 2017-06-03 |
* | Drop blockDim.y for the equivalent in_cols, and slightly improve naming (use ... | A. Unique TensorFlower | 2017-06-02 |
* | In the CUDA path of depthwise_conv2d, add a fast variant for forward convolut... | A. Unique TensorFlower | 2017-05-28 |
* | In the CUDA path of depthwise_conv2d, add __launch_bounds__ and use CUDA runt... | A. Unique TensorFlower | 2017-05-08 |
* | In the CUDA path of depthwise_conv2d, use compile-time constants if the filte... | A. Unique TensorFlower | 2017-05-03 |
* | Remove __launch_bounds__ from DepthwiseConv2dBackpropInputGPUKernelNHWC. | A. Unique TensorFlower | 2017-03-17 |
* | Add initial support for NCHW format for DepthwiseConv2D (and separable_conv2d). | Vijay Vasudevan | 2017-03-08 |
* | Improve speed of depthwise conv backward on GPU. | A. Unique TensorFlower | 2017-03-02 |
* | Set explicit launch bounds on DepthwiseConv2dBackpropInputGPUKernel(). | A. Unique TensorFlower | 2017-02-28 |
* | Merge changes from github. | Xiaoqiang Zheng | 2016-10-28 |
* | Update copyright for 3p/tf/core. | A. Unique TensorFlower | 2016-06-02 |
* | Add native depthwise_convolution op (backward pass) with GPU kernels. Now the | Jianmin Chen | 2016-03-04 |
* | Rollback of: Add native depthwise_convolution op (forward pass). | Jianmin Chen | 2016-02-25 |
* | Rollback of: Add native depthwise_convolution op (forward pass). | Vijay Vasudevan | 2016-02-25 |
* | Add native depthwise_convolution op (forward pass). | Jianmin Chen | 2016-02-25 |