Suffian Khan 46e0e4e69f Tune BiasGeluGradDx kernel in approximation mode to avoid tanh(...) on Rocm (#6239) ... * bias gelu grad use exp(...) instead * update cuda to rocm * missing semicolon * comment * remove dockerfile * missing factor of two		2021-01-02 08:54:16 -08:00
..
orttraining	Tune BiasGeluGradDx kernel in approximation mode to avoid tanh(...) on Rocm (#6239)	2021-01-02 08:54:16 -08:00
pytorch_frontend_examples	Enable running the mnist_training sample without cuda (#6085)	2020-12-15 17:06:54 -08:00
tools	Dockerfile to build onnxruntime with ROCm 4.0	2020-12-22 10:21:12 -08:00