Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add ONNX Runtime to ML Images #12

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add ONNX Runtime to ML Images #12

wants to merge 1 commit into from

Conversation

lreiher
Copy link
Member

@lreiher lreiher commented Nov 25, 2024

Installs ONNX Runtime with CUDA/TensorRT execution providers (onnxruntime-gpu; see here) to -ml images.

Closes #11.

@lreiher lreiher requested a review from jpbusch November 25, 2024 09:46
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Regenerate version tables including new column for ONNXRT.

# install ONNX Runtime
ARG ONNX_RUNTIME_VERSION
RUN if [[ -n $ONNX_RUNTIME_VERSION ]]; then \
pip3 install onnxruntime-gpu==${ONNX_RUNTIME_VERSION}; \
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Test ONNX RT, see #11 (comment).

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This sample script could be used for testing.

It currently fails with an error that libcudnn.so.9 is not found. According to the docs, the newest ONNX-RT version requires cuDNN 9.x. We are still on cuDNN 8.x though.

Downgrading to onnxruntime-gpu~=1.8.0 (for cuDNN 8.x), however, throws a similar error for libcublas, where we have v12 installed instead of v11.

2024-11-25 20:33:08.247359117 [E:onnxruntime:Default, provider_bridge_ort.cc:1745 TryGetProviderInfo_CUDA] /onnxruntime_src/onnxruntime/core/session/provider_bridge_ort.cc:1426 onnxruntime::Provider& onnxruntime::ProviderLibrary::Get() [ONNXRuntimeError] : 1 : FAIL : Failed to load library libonnxruntime_providers_cuda.so with error: libcublasLt.so.11: cannot open shared object file: No such file or directory

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Support for ONNX Runtime?
1 participant