Member-only story
Debugging the TensorFlow / Cuda error on AWS — ImportError: libcublas.so.9.0: cannot open shared object file
2 min readJun 14, 2020
Press enter or click to view image in full size![]()
Cause of error
This error is caused because mismatch in versions of tensorflow-gpu and CUDA. Every tensorflow-gpu lib is dependent on a very specific CUDA version.
Check our versions
Check tensorflow-gpu version :
pip list | grep tensorflow-gpuOur tensorflow-gpu version is 1.8.0.
Check CUDA version:
ls -l /usr/local/cudaOur cuda version is cuda-8.0.
Investigate issue
What are the compatible cuda versions for tensorflow :
Let’s refer to official TensorFlow page for the version compatibility.
Press enter or click to view image in full size![]()
So required dependencies for our tensorflow-gpu versions are:
So we have 2 options:
- Downgrade tensorflow-gpu to version 1.4.0 to match our system cuda version cuda-8.0
- Upgrade system cuda version to cuda-9.0 to match our…
