Help Center/ ModelArts/ Troubleshooting/ DevEnviron/ Code Running Failures/ cudaCheckError Occurs During Training
Updated on 2025-06-06 GMT+08:00

cudaCheckError Occurs During Training

Symptom

The following error occurs when the training code is run in a notebook instance:

cudaCheckError() failed : no kernel image is available for execution on the device

Possible Causes

Parameters arch and code in setup.py have not been set to match the GPU compute.

Solution

For GP Vnt1 GPUs, the GPU compute is -gencode arch=compute_70,code=[sm_70,compute_70]. Set the parameters in setup.py accordingly.