Hi Nvidia Team, Actually, I am working on registering a Plugin for an Operator(Einsum) which is not currently supported in TensorRT. So, instead of implementing a CUDA Kernel, I want to use the CuBLAS Library for Batch Matrix Multiplication. The Equations I want to implement is(from Einsum Operator): “ntg, ncg → nct” and " nct, ncp-> ntp" (for Batch Matrix Multiplication) Info about Einsum op: onnx/Operators.md at master · onnx/onnx · GitHub I needed a guidance in using CuBLAS Library for Batched Matrix Multiplication for the above two Ops. I am referring to the Available references(cuBLAS :: CUDA Toolkit Documentation, Pro Tip: cuBLAS Strided Batched Matrix Multiply | NVIDIA Developer Blog), but I am not getting how to use it for the above Equations. Can you please assist me for the same? Thanks in Advance, Darshan C G

[image] Batch Matrix Multiplication using CuBLAS GPU-Accelerated Libraries Hi Nvidia Team, Actually, I am working on registering a Plugin for an Operator(Einsum) which is not currently supported in TensorRT. So, instead of implementing a CUDA Kernel, I want to use the CuBLAS Library for Batch Matrix Multiplication. The Equations I want to implement is(from Einsum Operator): “ntg, ncg → nct” and " nct, ncp-> ntp"(for Batch Matrix Multiplication) Info about Einsum op: https://github.com/onnx/onnx/blob/master/docs/Operators.md#Einsum I needed a guidance in using CuBL…

Batch Matrix Multiplication using CuBLAS

Accelerated Computing GPU-Accelerated Libraries

Robert_Crovella February 19, 2021, 8:54pm 2

Topic		Replies	Views
Batch Matrix Multiplication using CuBLAS GPU-Accelerated Libraries tensorrt , cuda , kernel , c-plus-plus	17	4296	March 2, 2021
Pro Tip: cuBLAS Strided Batched Matrix Multiply Technical Blog	11	1307	February 16, 2018
Pro Tip: cuBLAS Strided Batched Matrix Multiply Technical Blog	0	446	November 1, 2021
Can cublasXt be used to process batched matmul? NVIDIA Nemotron cuda	0	54	September 11, 2024
cuBLAS matrix multiplication of a batch CUDA Programming and Performance	0	458	May 2, 2020
Issue with cuBLAS Batched Matrix Multiplication Result GPU-Accelerated Libraries cuda , pytorch	1	1055	March 22, 2021
SgemmBatched to multiply batched matrix and non batched matrix CUDA Programming and Performance	1	1141	April 16, 2015
Some Guidance on optimal approach to batch Matrix Multiply GPU-Accelerated Libraries cuda	0	609	August 11, 2020
cuBLAS convolution does not use Tensor Cores GPU-Accelerated Libraries cublas	6	2510	June 8, 2021
Matrix multiplication of many small-sized matrices CUDA Programming and Performance	3	1803	March 30, 2020

Batch Matrix Multiplication using CuBLAS

Related topics