Error installing tensorflow gpu ubuntu

General Discussion
,
#1

Hello, I am following the official guide on how to install Tensorflow using pip.
First I have detected what kind of GPU I am using on my Ubuntu 20 machine

$ lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GK107 [GeForce GT 640 OEM] (rev a1)

Then I downloaded the NVIDIA drivers from the official page, I have selected the one for the GeForce 600 series, specifically the 470.141.03 version.
The process of installing conda and creating a new environment went smoothly and also while installing tensorflow through pip I did not encounter any kind of error.
Unfortunately now that I am verifying if everything works correctly I get the following error

$ python3 -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"

2022-09-29 14:11:23.467663: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-09-29 14:11:23.467704: 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.
2022-09-29 14:11:23.520203: 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
2022-09-29 14:11:24.852647: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2022-09-29 14:11:24.852760: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2022-09-29 14:11:24.852779: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: 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.
2022-09-29 14:11:25.664198: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-09-29 14:11:25.664624: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.664706: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.664791: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.664867: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.664994: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.665069: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusolver.so.11'; dlerror: libcusolver.so.11: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.665151: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.665216: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory
2022-09-29 14:11:25.665229: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1934] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
tf.Tensor(215.02896, shape=(), dtype=float32)

This also happens if I try to verify the GPU setup. I didn’t find a solution on the wiki, does anyone know how to fix this?

#3

Have you followed the tested build configurations? Thank you.

#5

I was looking at the various combinations and I have a doubt, is there a way to find which version of cuda and cudnn are supported by my GPU?

#6

Looking at the hardware requirements it seems that the GPU must be compatible with Cuda 3.5 or higher.
This seems to be incompatible with my GPU looking at this guide. I think that for this reason is not possible to make tensorflow-gpu work on my computer.
I would still like to know if Tensorflow with only CPU compatibility can be successfully installed.

#7

I would still like to know if Tensorflow with only CPU compatibility can be successfully installed.

Yes, Tensorflow was installed successfully hence it returns tf.Tensor(215.02896, shape=(), dtype=float32) . Thank you.