Issue with installing tensorflow gpu

General Discussion
,
#1

I have this error while installing tensorflow 2.15 with cuda 12.2 and cudnn 2.8 can you tell me how to resolve this error

error
(tensorflow) C:\Windows\system32>python3 -c “import tensorflow as tf; print(tf.config.list_physical_devices(‘GPU’))”
2024-04-24 23:19:42.816954: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cudart64_101.dll’; dlerror: cudart64_101.dll not found
2024-04-24 23:19:42.817114: 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.
2024-04-24 23:19:44.455261: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library nvcuda.dll
2024-04-24 23:19:45.604509: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1716] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3050 Laptop GPU computeCapability: 8.6
coreClock: 1.5GHz coreCount: 16 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 178.84GiB/s
2024-04-24 23:19:45.604939: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cudart64_101.dll’; dlerror: cudart64_101.dll not found
2024-04-24 23:19:45.605968: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cublas64_10.dll’; dlerror: cublas64_10.dll not found
2024-04-24 23:19:45.607205: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cufft64_10.dll’; dlerror: cufft64_10.dll not found
2024-04-24 23:19:45.607564: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘curand64_10.dll’; dlerror: curand64_10.dll not found
2024-04-24 23:19:45.608007: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cusolver64_10.dll’; dlerror: cusolver64_10.dll not found
2024-04-24 23:19:45.608323: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cusparse64_10.dll’; dlerror: cusparse64_10.dll not found
2024-04-24 23:19:45.608604: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘cudnn64_7.dll’; dlerror: cudnn64_7.dll not found
2024-04-24 23:19:45.608804: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1753] 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 Install TensorFlow with pip for how to download and setup the required libraries for your platform.
Skipping registering GPU devices…
[]

#3

Welcome @Norhan_Reda_Abdelwah to the TensorFlow Forum,

It is clearly stated in the official TensorFlow documentation that:

TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows. Starting with TensorFlow 2.11, you will need to install TensorFlow in WSL2, or install tensorflow or tensorflow-cpu and, optionally, try the TensorFlow-DirectML-Plugin

Thus, you have two options in order to utilize your GPU with MS Windows:

  1. Install WSL2 and then install TensorFlow (whatever version you prefer),
  2. Install a compatible version of TensorFlow (<2.11)

I hope it helps!

#5

but I’m not installing it directly in windows I’m installing it with anaconda environment

#6

@Norhan_Reda_Abdelwah I am sorry but unfortunately, according to the documentation you can not install TensorFlow version greater than 2.10 if you wish to utilise your GPU in native Windows regardless if TensorFlow is installed within a conda virtual environment.