onnxruntime
9174cbe3 - Optimize CUDA Kernel for 3D and 4D Transpose (#8928)

Commit
4 years ago
Optimize CUDA Kernel for 3D and 4D Transpose (#8928) * Optimize Transpose120 and Transpose102 * Generalize Transpose0123 for more input shapes * Add Transpose3D test cases * update rocm kernel
Author
Parents
Loading