Aleksei Zinovev

Also, there could be a problem with the drivers for the GPU card or memory allocation https://forums.developer.nvidia.com/t/error-failed-to-run-cublas-routine-cublas-status-execution-failed/164278 https://github.com/matterport/Mask_RCNN/issues/2510

Also, if you experimented with many CUDA versions, probably, you need to check twice the environment variables; the LIBRARY_PATH or PATH could contain the path to another CUDA version (it...

We could provide such kind of options in the future release This should be solved in the #540

Yeah, there are a few ways to integrate Torch in Kotlin 1) JNI 2) JavaCPP 3) Pure PyTorch Java API with IValue for inference only Hope that in 2022 KotlinDL...

Thanks, @elonzh a good point to revisit the API

Mr. Tomasz, I'm also faced with a similar issue with the API (missed method on operand level, only on builder). Do you have a chance to experiment on the previous...

Hi @brunoteo thanks for the contribution! We need approximately 2–4 weeks to test your solution and make a decision about including this feature in 0.14 release or not. After that,...

@Burtan if it's actual, and you keep the example with the invalid input, could you please share the code/data with us?

@Jolanrensen please check

Probably close, if won't fix?