How to resolve when ML Skill fails with CUDNN_STATUS_EXECUTION_FAILED?
Resolution:
- Make sure the GPU that is allocated to the machine has enough resources. Read more here: Training On GPU or On CPU (Make sure to be on the correct version of the documentation).
- Check /var/log/messages for any Nvidia-related errors.
- Ensure the below pre-requisites are met:
- For AI Center installation, it is a pre-requirement that the node has NVIDIA driver version 450.51.06 installed, as well as Nvidia-container-runtime.
- "nvidia-smi" command should list GPU details.
- In case nothing works, generate a support bundle by following the procedure at UiPath Support Bundle and contact UiPath Technical Support.