How do I configure my jupyter notebook so that it uses the available GPU while working with keras?

1
votes

I searched for solutions and installed tensorflow-gpu with pip.

tf.config.list_physical_devices('GPU')

This code returns an empty list. []

You can see I have 2 GPUs but none of them are being used when I am doing image processing(CNN) with keras.

I am new to things so don't understand exactly what is wrong. Please help me configure so that I can use my GPU for processing. I am working with Windows 10 64-bit, Python-3.8.7. As suggested, I tried importing tensorflow on python terminal, and got the following error:

import tensorflow as tf

2021-02-13 22:52:17.253841: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2021-02-13 22:52:17.266384: 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.

Followed by this when I tried listing the GPUs: tf.config.list_physical_devices('GPU')

2021-02-13 22:57:17.390319: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set 2021-02-13 22:57:17.749790: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library nvcuda.dll 2021-02-13 22:57:18.937838: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: pciBusID: 0000:01:00.0 name: GeForce MX150 computeCapability: 6.1 coreClock: 1.5315GHz coreCount: 3 deviceMemorySize: 2.00GiB deviceMemoryBandwidth: 44.76GiB/s 2021-02-13 22:57:18.966071: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2021-02-13 22:57:18.974209: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cublas64_11.dll'; dlerror: cublas64_11.dll not found 2021-02-13 22:57:18.981154: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cublasLt64_11.dll'; dlerror: cublasLt64_11.dll not found 2021-02-13 22:57:18.988826: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cufft64_10.dll'; dlerror: cufft64_10.dll not found 2021-02-13 22:57:18.996411: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'curand64_10.dll'; dlerror: curand64_10.dll not found 2021-02-13 22:57:19.002563: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cusolver64_10.dll'; dlerror: cusolver64_10.dll not found 2021-02-13 22:57:19.009636: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cusparse64_11.dll'; dlerror: cusparse64_11.dll not found 2021-02-13 22:57:19.018025: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudnn64_8.dll'; dlerror: cudnn64_8.dll not found 2021-02-13 22:57:19.025064: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1757] 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... []

1
pythontensorflowkerasjupyter-notebookgpu-shared-memory
Have you installed CUDA Toolkit and cuDNN ? There are countless tutorials for that on the internet... – Omar Aflak
will try, thankyou! – ANKIT CHOUDHARY
Run your code in a python shell (not in jupyter) and you will see a log that tells you everything on why the GPU is not detected. You could include it in your question (as text, not as image). – Dr. Snoopy
If you import tf on terminal, you should see I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll otherwise there something wrong in your set up. – M.Innat
I got an error message while importing tf on python terminal, mentioned it in the question. – ANKIT CHOUDHARY

1 Answers

1
votes

Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found This error message says Tensorflow-gpu requires CUDA 11

Could not load dynamic library 'cudnn64_8.dll'; dlerror: cudnn64_8.dll not found This error message says cuDNN 8 is required to run Tensorflow-gpu version.

Install right version CUDA and cuDNN and follow the instructions mentioned here to setup GPU support on Windows OS.