-
Notifications
You must be signed in to change notification settings - Fork 75.3k
2.10.0 cuBLAS error #57663
Copy link
Copy link
Closed
Labels
TF 2.10staleThis label marks the issue/pr stale - to be closed automatically if no activityThis label marks the issue/pr stale - to be closed automatically if no activitystat:awaiting responseStatus - Awaiting response from authorStatus - Awaiting response from authorsubtype: ubuntu/linuxUbuntu/Linux Build/Installation IssuesUbuntu/Linux Build/Installation Issuestype:build/installBuild and install issuesBuild and install issues
Description
Click to expand!
Issue Type
Build/Install
Source
source
Tensorflow Version
2.10.0
Custom Code
No
OS Platform and Distribution
Ubuntu 7.5.0-3ubuntu1~18.04
Mobile device
No response
Python version
3.8.0
Bazel version
No response
GCC/Compiler version
gcc 7.5.0
CUDA/cuDNN version
NVIDIA-SMI 450.102.04 Driver Version: 450.102.04 CUDA Version: 11.0
GPU model and memory
No response
Current Behaviour?
I use `pip` to install tensorflow-2.10.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl`. After installation, when I import tensorflow, there is an error:
`E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered.
I noticed this is an error that occurred before and has an issue closed. But I install the latest 2.10.0 version tf and still meet this error. hope you can help me figure it out. thanks!Standalone code to reproduce the issue
pip install tensorflow-2.10.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whlRelevant log output
In [1]: import tensorflow as tf
2022-09-11 03:59:51.293241: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2022-09-11 03:59:51.436444: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2022-09-11 03:59:51.474370: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered</details>
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
TF 2.10staleThis label marks the issue/pr stale - to be closed automatically if no activityThis label marks the issue/pr stale - to be closed automatically if no activitystat:awaiting responseStatus - Awaiting response from authorStatus - Awaiting response from authorsubtype: ubuntu/linuxUbuntu/Linux Build/Installation IssuesUbuntu/Linux Build/Installation Issuestype:build/installBuild and install issuesBuild and install issues