Ubuntu 21.08 TF GPU problem

Mr_Sweaty · February 13, 2023, 2:00am

I have a problem with using the GPU inside Pycharm. I followed the guide on tensorflow to install. With all the correct extensions, on a freshly installed version of Ubuntu 21.08
Python 3.9.16
CUDA version: 11.2
CUDNN version: 8

When i do the following code inside pycharm i get:

import tensorflow as tf

print('TensorFlow version:',tf.__version__)
physical_devices = tf.config.list_physical_devices()
for dev in physical_devices:
    print(dev)

sys_details = tf.sysconfig.get_build_info()
cuda_version = sys_details["cuda_version"]
print("CUDA version:",cuda_version)
cudnn_version = sys_details["cudnn_version"]
print("CUDNN version:",cudnn_version)

I get:

TensorFlow version: 2.11.0
2023-02-10 14:33:29.539642: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-02-10 14:33:29.539993: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540081: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540156: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540229: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540302: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540374: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540445: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540517: W tensorflow/compiler/xla/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
2023-02-10 14:33:29.540532: 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...
PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU')
CUDA version: 11.2
CUDNN version: 8

I’ve run the

mkdir -p $CONDA_PREFIX/etc/conda/activate.d
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh

inside the Ubuntu Terminal, while inside the tf conda environment.

I don’t know what to do at this point. Please help

chunduriv · February 13, 2023, 5:46am

@Mr_Sweaty,

Welcome to the Tensorflow Forum!

CUDNN version: 8

According to tested build configurations please upgrade cuDNN to 8.1.0 and try again?

If the issue still continues, try to point your bin file from CUDA installation in your environment path as suggests in this thread .

Thank you!

Mr_Sweaty · February 13, 2023, 11:51am

Thanks. I’ve tried with both 8.1.0, 8.1.1, 8.1.2. I think it has something to do with the LD_Library path. But the tensorflow.org step by step does not cover this issue. In step 4. It says:
Configure the system paths. You can do it with the following command every time you start a new terminal after activating your conda environment.

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/

For your convenience it is recommended that you automate it with the following commands. The system paths will be automatically configured when you activate this conda environment.

mkdir -p $CONDA_PREFIX/etc/conda/activate.decho 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/' > $CONDA_PREFIX/etc/conda/activate.d

But the LD_LIBRARY must contain the libnvinfer.so.7. But it does not. the libninfer.so.7 does not come from the cuda or cudnn either (As far as I can tell, by rumaging through the files)

tensorflow/compiler/xla/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; LD_LIBRARY_PATH: :/home/victor/miniconda3/lib/:/home/victor/miniconda3/envs/tf/lib/

I then tried to find a program that actually had these libs. The TensorRT has these libs. So I took them and copied to the path. This however did not resolve the problem either.

chunduriv · February 13, 2023, 1:19pm

But the LD_LIBRARY must contain the libnvinfer.so.7.But it does not.

As an optional. TensorRT to improve latency and throughput for inference.

tensorflow/compiler/xla/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; LD_LIBRARY_PATH: :/home/victor/miniconda3/lib/:/home/victor/miniconda3/envs/tf/lib/

According to #57671, above warnings are caused by outdated TensorRT dependency in the pypi wheel for Linux, which will be fixed in TF 2.12.

Thank you!

Mr_Sweaty · February 13, 2023, 2:09pm

I’m not using TensorRT.
Anyways. When in Pycharm terminal, and i run: python3 -c “import tensorflow as tf; print(tf.config.list_physical_devices(‘GPU’))” i get:

(tf) victor@victor-ThinkPad-P53:~/PycharmProjects/Collect$ python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
2023-02-13 14:52:45.667286: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-02-13 14:52:46.422943: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvrtc.so.11.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/home/victor/miniconda3/lib/:/home/victor/miniconda3/envs/tf/lib/
2023-02-13 14:52:46.423087: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvrtc.so.11.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/home/victor/miniconda3/lib/:/home/victor/miniconda3/envs/tf/lib/
2023-02-13 14:52:46.423095: 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.
2023-02-13 14:52:48.604221: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-02-13 14:52:48.720010: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-02-13 14:52:48.720166: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
(tf) victor@victor-ThinkPad-P53:~/PycharmProjects/Collect$

So the pycharm terminal looking at the conda environment can see the GPU.
However, when executing the following code from a .py file in the same conda environment i get:

import tensorflow as tf

print('TensorFlow version:',tf.__version__)
physical_devices = tf.config.list_physical_devices()
for dev in physical_devices:
    print(dev)

sys_details = tf.sysconfig.get_build_info()
cuda_version = sys_details["cuda_version"]
print("CUDA version:",cuda_version)
cudnn_version = sys_details["cudnn_version"]
print("CUDNN version:",cudnn_version)
print(tf.config.list_physical_devices('GPU'))

/home/victor/miniconda3/envs/tf/bin/python /home/victor/PycharmProjects/Collect/Code/import tester.py 
2023-02-13 15:06:30.402827: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-02-13 15:06:30.488652: W tensorflow/compiler/xla/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
2023-02-13 15:06:30.488664: I tensorflow/compiler/xla/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2023-02-13 15:06:31.007126: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvrtc.so.11.1: cannot open shared object file: No such file or directory
2023-02-13 15:06:31.007238: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvrtc.so.11.1: cannot open shared object file: No such file or directory
2023-02-13 15:06:31.007245: 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.
TensorFlow version: 2.11.0
PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU')
CUDA version: 11.2
CUDNN version: 8
[]
2023-02-13 15:06:33.062143: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-02-13 15:06:33.062287: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062319: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062347: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062373: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062400: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062426: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062451: W tensorflow/compiler/xla/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
2023-02-13 15:06:33.062547: 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...

Process finished with exit code 0

It skips looking for the GPU, because something is missing. But whatever is missing is quite hard to find, because of the bug that apparently exists.

What do I do?

Suvrodeep_Ghosh · March 22, 2023, 1:19am

@Mr_Sweaty I have the exact same problem. First things first libinfer.so.7 does not impact using TF on the GPU so I am not bothered about it at the moment. From what I can tell it helps in inference.

Coming to the issue of not being able to load libcudart.so.11. Using Python 3.9 with TensorFlow=2.10.1. I even launched an interactive prompt with ipython and there I can see the GPU but not when I launch the python console from PyCharm or when running a script from within PyCharm. I tried adding the LD_LIBRARY_PATH value, which actually contains libcudart.so.11, to the interpreter paths but still the issue persists.

Have you been able to fins a solution to this ?

Best regards and thanks in advance!

Suvrodeep_Ghosh · March 22, 2023, 3:47am

(ml_dev) ghubro@ghubro-ml-system:~/GitHub/Playing-with-Neural-Networks$ ipython
Python 3.9.16 (main, Mar  8 2023, 14:00:05) 
Type 'copyright', 'credits' or 'license' for more information
IPython 8.10.0 -- An enhanced Interactive Python. Type '?' for help.

In [1]: import tensorflow as tf
2023-03-22 00:09:49.916218: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-22 00:09:50.001479: 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
2023-03-22 00:09:50.335880: 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; LD_LIBRARY_PATH: :/home/ghubro/anaconda3/envs/ml_dev/lib/
2023-03-22 00:09:50.335921: 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; LD_LIBRARY_PATH: :/home/ghubro/anaconda3/envs/ml_dev/lib/
2023-03-22 00:09:50.335927: 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.

In [2]: print(f"GPUs:{len(tf.config.list_logical_devices(device_type='GPU'))}")
2023-03-22 00:11:27.010999: 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
2023-03-22 00:11:27.013613: 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
2023-03-22 00:11:27.013735: 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
2023-03-22 00:11:27.014122: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-22 00:11:27.014438: 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
2023-03-22 00:11:27.014549: 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
2023-03-22 00:11:27.014646: 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
2023-03-22 00:11:27.287663: 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
2023-03-22 00:11:27.287804: 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
2023-03-22 00:11:27.287902: 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
2023-03-22 00:11:27.287993: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1616] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 5422 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3060 Ti, pci bus id: 0000:07:00.0, compute capability: 8.6
GPUs:1

In [3]:

Here I can see the GPU but below from inside the python console I cannot:

home/ghubro/anaconda3/envs/ml_dev/bin/python /home/ghubro/.local/share/JetBrains/Toolbox/apps/PyCharm-C/ch-0/222.4345.23/plugins/python-ce/helpers/pydev/pydevconsole.py --mode=client --host=127.0.0.1 --port=43177 
import sys; print('Python %s on %s' % (sys.version, sys.platform))
sys.path.extend(['/home/ghubro/GitHub/Playing-with-Neural-Networks'])
Python 3.9.16 (main, Mar  8 2023, 14:00:05) 
Type 'copyright', 'credits' or 'license' for more information
IPython 8.10.0 -- An enhanced Interactive Python. Type '?' for help.
PyDev console: using IPython 8.10.0
Python 3.9.16 (main, Mar  8 2023, 14:00:05) 
[GCC 11.2.0] on linux
import tensorflow as tf
2023-03-22 00:12:48.430031: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-22 00:12:48.505657: 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
2023-03-22 00:12:48.505670: 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.
2023-03-22 00:12:48.519403: 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
2023-03-22 00:12:48.867585: 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
2023-03-22 00:12:48.867619: 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
2023-03-22 00:12:48.867623: 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.
print(f"GPUs:{len(tf.config.list_logical_devices(device_type='GPU'))}")
2023-03-22 00:13:25.944361: 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
2023-03-22 00:13:25.944566: 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
2023-03-22 00:13:25.944606: 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
2023-03-22 00:13:25.944639: 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
2023-03-22 00:13:25.944670: 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
2023-03-22 00:13:25.944701: 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
2023-03-22 00:13:25.944732: 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
2023-03-22 00:13:25.944762: 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
2023-03-22 00:13:25.944791: 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
2023-03-22 00:13:25.944798: 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...
2023-03-22 00:13:25.945163: 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 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
GPUs:0


Tried to add the path containing libcudart.so.11.0 here but to no avail:

![Screenshot from 2023-03-22 00-14-47|627x494](upload://2e0oryb9Sg14sWBS3PdC54OgphC.png)