aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/core/kernels/conv_ops_gpu.h
Commit message (Expand)AuthorAge
* Do not segfault in Conv2d/3d if cuDNN version is too low.Gravatar Reed Wanderman-Milne2018-09-17
* Add data format as a parameter in ConvParameters to support NHWC formatGravatar Smit Hinsu2018-07-24
* Merge changes from github.Gravatar Yifei Feng2018-04-23
* Use tensorflow::se instead of perftools::gputools for StreamExecutor.Gravatar Justin Lebar2018-04-23
* Eliminate the guard around Winograd non-fused convolutions with cudnn7.Gravatar A. Unique TensorFlower2018-04-20
* Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/core/kernels.Gravatar A. Unique TensorFlower2018-01-26
* Merge changes from github.Gravatar Patrick Nguyen2017-12-28
* Add native dilated support for conv2d and its gradients in cudnn v>=6.Gravatar Yangzihao Wang2017-11-29
* Add int8 version of fused_conv2d_bias_activation operator for the forward phase,Gravatar A. Unique TensorFlower2017-09-06
* Automated g4 rollback of changelist 166276461Gravatar A. Unique TensorFlower2017-08-24
* Add int8 version of fused_conv2d_bias_activation operator for the forward phase,Gravatar A. Unique TensorFlower2017-08-23
* Return error status for invalid argument for ScratchAllocator.Gravatar Yangzihao Wang2017-08-10
* Let GetBlasGemmAlgorithms() always return true.Gravatar Yangzihao Wang2017-07-21
* Add fused_conv2d_bias_activation operator for the forward phase.Gravatar Yangzihao Wang2017-07-20
* Automated g4 rollback of changelist 162423171Gravatar A. Unique TensorFlower2017-07-18
* Add autotuning code for matmul operator.Gravatar Yangzihao Wang2017-07-18
* Set winograd nofused flag to be true by default.Gravatar Yangzihao Wang2017-05-28
* Add a multi-pass mode for autotuning. It tests the autotune multiple times,Gravatar Xiaoqiang Zheng2017-03-21
* Added data type info to conv autotune parameters.Gravatar Yangzihao Wang2017-03-17
* Generalize autotuning to 3d ops.Gravatar A. Unique TensorFlower2017-03-15
* Change Autotune to the global scope. Switch to hash table for the autotuneGravatar Xiaoqiang Zheng2016-10-10
* Relax static shape information requirements for depthwise and separable conv2d.Gravatar A. Unique TensorFlower2016-08-17
* Improve convolution autotune process. The max batch size VGG model can handleGravatar Xiaoqiang Zheng2016-06-21
* Update copyright for 3p/tf/core.Gravatar A. Unique TensorFlower2016-06-02
* In case of memory allocation failure, return an error status to streamGravatar Xiaoqiang Zheng2016-05-25
* Change the ConvParameters comparison function to use std::tuple to avoidGravatar Xiaoqiang Zheng2016-05-23
* Add the autotune version for the backward passes. It is currentlyGravatar Xiaoqiang Zheng2016-05-11
* Use the untemplated version of OpKernelContext::op_device_context() soGravatar Josh Levenberg2016-03-11
* Merge changes from GitHub, clean up linter errors, fix dependencies test.Gravatar Martin Wicke2016-02-07
* Disable tensor tracking when only one GPU stream is used.Gravatar Xiaoqiang Zheng2016-02-02
* Many tensorflow/core build clean ups.Gravatar Josh Levenberg2016-01-20
* Replacing reference 'names' variable with 'example_names' variable.Gravatar A. Unique TensorFlower2016-01-20
* Remove an unnecessary warning.Gravatar Xiaoqiang Zheng2016-01-14
* The default memory manager would delay 10s for any memory failure. This is no...Gravatar A. Unique TensorFlower2016-01-07
* Improve TensorFlow convnet performance.Gravatar A. Unique TensorFlower2015-12-17
* TensorFlow: Upstream changes to git.Gravatar Vijay Vasudevan2015-12-08
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-18 10:44:32 +0000