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
Kernel crash when using TensorFlow/PyTorch? #8
Comments
@mthiboust See the CUDA Version Matrix for PyTorch/TensorFlow compatibility. |
Thanks @benz0li for your help. Your images haven't solved my problem. In fact, the root cause may be on Keras side because I have the same problem with tensorflow official image (cf keras-team/keras#19601). |
@mthiboust I cannot reproduce with image docker run --rm -ti glcr.b-data.ch/jupyterlab/cuda/python/scipy:3.12.3 bash
In the Container: pip install keras pandas tensorflow
nano test.py 👉 Python code from keras-team/keras#19601 (comment) python test.py
Code run on CPU (Intel(R) Xeon(R) Silver 4210R) / GPU (Quadro RTX 4000, Compute Capability 7.5) / Ubuntu 22.04 (Container) with Keras 3.3.3, Numpy 1.26.4 and Tensorflow 2.16.1. |
Thanks for testing it! Like you, I do not have the issue when using the GPU. The bug only happens on CPU. I am curious to know if you run into the same bug on your CPU without CUDA |
Checking right now... |
@mthiboust No. I cannot reproduce with image docker run --rm -ti glcr.b-data.ch/jupyterlab/python/scipy:3.12.3 bash
In the Container: pip install keras pandas tensorflow
nano test.py 👉 Python code from keras-team/keras#19601 (comment) python test.py
Code run on CPU (Intel(R) Xeon(R) Silver 4210R) / Ubuntu 22.04 (Container) with Keras 3.3.3, Numpy 1.26.4 and Tensorflow 2.16.1. |
Thanks again @benz0li! I'll have a new look at it next week. Hopefully I'll have new ideas by then! |
Closing because I cannot reproduce [on my machine]. |
@mthiboust Could you test with my/b-data's images?
glcr.b-data.ch/jupyterlab/cuda/python/scipy:3.11.9
nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
Install TensorFlow:
pip install tensorflow==2.14.1
Install PyTorch:
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
glcr.b-data.ch/jupyterlab/cuda/python/scipy:3.12.3
nvidia/cuda:12.4.1-devel-ubuntu22.04
Install TensorFlow:
pip install tensorflow
Install PyTorch:
pip install torch torchvision torchaudio
Cross reference:
The text was updated successfully, but these errors were encountered: