Hello guys, sorry for asking this simple question but why won’t tensorflow detect my GPU I have the correct versions for python, cuda, cudnn, and tensorflow and I think I’m stumped after 3 days. I hope someone can help. Here is a summary of installed packages I used in Anaconda Powershell Prompt.
(tf) PS C:\WINDOWS\system32> conda list cudatoolkit
# packages in environment at C:\ProgramData\anaconda3\envs\tf:
#
# Name Version Build Channel
cudatoolkit 11.2.2 h7d7167e_13 conda-forge
(tf) PS C:\WINDOWS\system32> conda list cudnn
# packages in environment at C:\ProgramData\anaconda3\envs\tf:
#
# Name Version Build Channel
cudnn 8.1.0.77 h3e0f4f4_0 conda-forge
(tf) PS C:\WINDOWS\system32> conda list tensorflow
# packages in environment at C:\ProgramData\anaconda3\envs\tf:
#
# Name Version Build Channel
tensorflow 2.10.0 pypi_0 pypi
tensorflow-estimator 2.10.0 pypi_0 pypi
(tf) PS C:\WINDOWS\system32> conda list python
# packages in environment at C:\ProgramData\anaconda3\envs\tf:
#
# Name Version Build Channel
python 3.9.19 h1aa4202_0
(tf) PS C:\WINDOWS\system32> python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
[]
(tf) PS C:\WINDOWS\system32> python -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"
tf.Tensor(-1392.3337, shape=(), dtype=float32)