Still get "THCudaBlas_Sgemm is undefined" error with Pytorch1.8+Cuda10.2 on Jetson AGX about dcnv2_latest HOT 9 CLOSED

Sorry, I can not reproduce. both pytorch 1.7 and 1.8 OK for me.

from dcnv2_latest.

nvcr.io/nvidia/pytorch:20.12-py3 this docker image could reproduce this error.

from dcnv2_latest.

@jinfagang I also get the same error with 1.8 recently, but works when downgrading to 1.7

from dcnv2_latest.

@jinfagang Same error with pytorch1.8

from dcnv2_latest.

@austinmw @McDifference That is weired. I tried pytorch 1.8 and it worked on my side. You can see my 2 different computer output message. Maybe 1.8.1 or 1.9 might help?

from dcnv2_latest.

@jinfagang Thanks a lot for your efforts. I have the same error with PyTorch 1.8.1 & cuda_11.1 on an RTX 3090.

from dcnv2_latest.

I also got the same problem! For the newest version of Pytorch (1.8.1+), they have moved away the THCBlas (pytorch/pytorch#49725) so the THCudaBlas_SgemmBatched, THCudaBlas_Sgemm cannot use anymore! I try the cuda code from this repository and compiled it successfully with PyTorch 1.8.1 & cuda_11.1 on an RTX 3070.

from dcnv2_latest.

I have updated and removed THCudaBlas_SgemmBatched

from dcnv2_latest.

I have updated and removed THCudaBlas_SgemmBatched

Hallo,
Can you share how you did that?

from dcnv2_latest.