search

microsoft / onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
https://onnxruntime.ai
MIT License
14.55k stars 2.91k forks source link

Does DML need a shrinkage ? #8189

Closed znsoftm closed 1 year ago

znsoftm commented 3 years ago

We know the fact that we could call run_option.AddConfigEntry(kOrtRunOptionsConfigEnableMemoryArenaShrinkage, "gpu:0") with CUDA. So do we need to call the same function for DML ?

When I add the line into a piece of code with DML, it crashed.

stale[bot] commented 2 years ago

This issue has been automatically marked as stale due to inactivity and will be closed in 7 days if no further activity occurs. If further support is needed, please provide an update and/or more details.