You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for your CUDA-INT8-GEMM project on GitHub. I have been perplexed by how to perform integer matrix operations using WMMA, and your project has been immensely helpful.
When I compiled with int8_t as the output type, everything ran smoothly; however, when I tried to set the output type to int, the compiler indicated that there seemed to be no overloading for int parameters, which puzzled me because gemmi8.cuh does contain such overloading.
I have tried various methods but cannot resolve this issue, so I am seeking guidance on how to use int as the output type correctly. If you have the time to respond to my question, I would be delighted.
The text was updated successfully, but these errors were encountered:
Thank you for your CUDA-INT8-GEMM project on GitHub. I have been perplexed by how to perform integer matrix operations using WMMA, and your project has been immensely helpful.
When I compiled with int8_t as the output type, everything ran smoothly; however, when I tried to set the output type to int, the compiler indicated that there seemed to be no overloading for int parameters, which puzzled me because gemmi8.cuh does contain such overloading.
I have tried various methods but cannot resolve this issue, so I am seeking guidance on how to use int as the output type correctly. If you have the time to respond to my question, I would be delighted.
The text was updated successfully, but these errors were encountered: