Commit message (Expand) | Author | Age | |
---|---|---|---|
* | Cudnn RNN v2 kernels with autotune capability | 2018-04-25 | |
* | Add new env-var TF_FP16_CONV_MODE. | 2017-11-03 | |
* | Add a flag to set whether to skip using cudnn for 1x1 filter. | 2017-08-09 | |
* | Use Cudnn for avgpool by default. | 2017-05-22 | |
* | Adding an environmental variable to control the cuda host memory limit. | 2016-10-12 | |
* | Adding an environment variable to control whether to use Cudnn with AvgPool: | 2016-07-11 | |
* | Update copyright for 3p/tf/core. | 2016-06-02 | |
* | Turn on CUDNN autotune by default. | 2016-05-24 | |
* | Adding autotune to the Cudnn conv algorithm selection. | 2016-05-09 | |
* | Global search & replace to move to the new location for | 2016-01-26 | |
* | TensorFlow: Improve performance of Alexnet | 2015-11-20 | |
* | TensorFlow: Initial commit of TensorFlow library. | 2015-11-06 |