Open srijanie03 opened 1 month ago
This typically happens when you have both onnxruntime as well as onnxruntime-training installed in the same env.
To resolve, uninstall all onnxruntime packages
pip uninstall -y onnxruntime onnxruntime-training-cpu
Confirm that you have no onnxruntime packages installed (use pip freeze to check any package with the prefix onnxruntime-).
And then, reinstall only onnxruntime-training-cpu.
Would I need to do a build from the start, or can I use the wheel file generated during the previous build and install onnxruntime-training as I did previously with:
python -m pip install build/Windows/RelWithDebInfo/RelWithDebInfo/dist/onnxruntime_training-1.19.0+cpu-cp312-cp312-win_amd64.whl
It seems to be working, thank you! I just continued with the original build - uninstalled all and installed only onnxruntime-training again.
Any reason why the error popped up all of a sudden? Why was it working okay earlier with both libraries coexisting?
onnxruntime-training contains everything onnxruntime has but it is released on the smaller set of os and versions.
This issue has been automatically marked as stale due to inactivity and will be closed in 30 days if no further activity occurs. If further support is needed, please provide an update and/or more details.
Describe the issue
I had onnxruntime.training working correctly on Windows x64. Out of nowhere, I get this error message:
The build was completed correctly. I have double-checked the msvcp140.dll and "vcruntime140.dll" files. I have updated vs2015_runtime, and have the latest package of vc++ running. What could be causing this sudden issue?
To reproduce
import onnxruntime.training
Urgency
Very urgent
ONNX Runtime Installation
Built from Source
ONNX Runtime Version or Commit ID
1.18.1
PyTorch Version
2.4.0
Execution Provider
Default CPU
Execution Provider Library Version
No response