profile
viewpoint

Ask questionsUsing tensorflow gpu 2.1 with Cuda 10.2

  • OS Platform and Distribution: Linux Ubuntu 16.04
  • TensorFlow installed from: pip
  • TensorFlow version: 2.1.0rc0
  • Python version: 3.6.8
  • Installed using virtualenv? pip? conda?: pip
  • CUDA/cuDNN version: 10.2
  • GPU model and memory: Quadro P5000, 16GB

Describe the problem

I want to use tensorflow-gpu==2.1.0rc0 with cuda 10.2 and it seems that it can't work right now. When I use tensorflow-gpu=2.0.0 it works perfectly fine.

Provide the exact sequence of commands / steps that you executed before running into the problem

mkdir tests2 &&\
cd tests2 &&\
virtualenv -p /usr/bin/python3.6 venv &&\
source venv/bin/activate &&\
pip install tensorflow-gpu==2.1.0rc0 &&\
python -c 'import tensorflow'

Which gives the following warnings:

2019-12-02 15:23:46.869198: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcudart.so.10.1'; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2019-12-02 15:23:46.869227: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2019-12-02 15:23:47.516321: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2019-12-02 15:23:47.516433: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2019-12-02 15:23:47.516449: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

Any other info / logs When I do locate libcudart.so, I get the following:

/usr/local/cuda-10.0/doc/man/man7/libcudart.so.7
/usr/local/cuda-10.0/targets/x86_64-linux/lib/libcudart.so
/usr/local/cuda-10.0/targets/x86_64-linux/lib/libcudart.so.10.0
/usr/local/cuda-10.0/targets/x86_64-linux/lib/libcudart.so.10.0.130
/usr/local/cuda-10.2/doc/man/man7/libcudart.so.7
/usr/local/cuda-10.2/targets/x86_64-linux/lib/libcudart.so
/usr/local/cuda-10.2/targets/x86_64-linux/lib/libcudart.so.10.2
/usr/local/cuda-10.2/targets/x86_64-linux/lib/libcudart.so.10.2.89

locate libnvinfer_plugin.so is empty.

tensorflow/tensorflow

Answer questions EwoutH

@zaccharieramzi They got Cuda 10.1 working in another thread: https://github.com/tensorflow/tensorflow/issues/34429#issuecomment-562961299. Could you verify?

useful!

Related questions

ModuleNotFoundError: No module named 'tensorflow.contrib' hot 8
Error occurred when finalizing GeneratorDataset iterator hot 6
ModuleNotFoundError: No module named 'tensorflow.contrib'
When importing TensorFlow, error loading Hadoop
tf.keras.layers.Conv1DTranspose ?
tensorflow-gpu CUPTI errors hot 4
[TF 2.0] tf.keras.optimizers.Adam hot 4
Lossy conversion from float32 to uint8. Range [0, 1]. Convert image to uint8 prior to saving to suppress this warning. hot 4
TF2.0 AutoGraph issue hot 4
Tf.Keras metrics issue hot 4
Error loading tensorflow hot 4
module 'tensorflow' has no attribute 'ConfigProto' hot 4
TF 2.0 'Tensor' object has no attribute 'numpy' while using .numpy() although eager execution enabled by default hot 4
ModuleNotFoundError: No module named 'tensorflow.examples.tutorials' hot 4
AttributeError: module 'tensorflow.python.framework.op_def_registry' has no attribute 'register_op_list' hot 4
Github User Rank List