0

I got into Deep Learning and Computer Vision through a Youtube Video and started programming and installed TensorFlow and TensorFlow GPU (also installed cuda and cudnn). To check if it was using my GPU I googled and saw this report: Check whether Tensorflow is running on GPU and ran the code. Here is my output:

import tensorflow as tf 2021-07-03 10:53:12.843163: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudart64_110.dll if tf.test.gpu_device_name(): ... print('Default GPU Device: {}'.format(tf.test.gpu_device_name())) ... else: ... print("Please install GPU version of TF") ... 2021-07-03 10:53:36.416599: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX AVX2 To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2021-07-03 10:53:36.421369: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library nvcuda.dll 2021-07-03 10:53:36.461645: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1733] Found device 0 with properties: pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 2060 computeCapability: 7.5 coreClock: 1.2GHz coreCount: 30 deviceMemorySize: 6.00GiB deviceMemoryBandwidth: 245.91GiB/s 2021-07-03 10:53:36.461807: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudart64_110.dll 2021-07-03 10:53:36.465569: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cublas64_11.dll'; dlerror: cublas64_11.dll not found 2021-07-03 10:53:36.467869: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cublasLt64_11.dll'; dlerror: cublasLt64_11.dll not found 2021-07-03 10:53:36.470114: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cufft64_10.dll'; dlerror: cufft64_10.dll not found 2021-07-03 10:53:36.472643: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'curand64_10.dll'; dlerror: curand64_10.dll not found 2021-07-03 10:53:36.474775: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cusolver64_11.dll'; dlerror: cusolver64_11.dll not found 2021-07-03 10:53:36.476926: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cusparse64_11.dll'; dlerror: cusparse64_11.dll not found 2021-07-03 10:53:36.478093: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudnn64_8.dll 2021-07-03 10:53:36.478147: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1766] 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... 2021-07-03 10:53:36.591753: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1258] Device interconnect StreamExecutor with strength 1 edge matrix: 2021-07-03 10:53:36.591877: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1264] 0 2021-07-03 10:53:36.593445: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1277] 0: N Please install GPU version of TF

As I mentioned earlier, I already installed it. How can I solve this problem?

Thilo Rabens
  • 1
  • 1
  • The reason is explained in that error message: you don't have the CUDA libraries in your PATH, e.g. "**Could not load dynamic library 'cublas64_11.dll'; dlerror: cublas64_11.dll not found**" – Dai Jul 03 '21 at 09:37
  • Did you try following the link in the error message and following the steps there? – Almog-at-Nailo Jul 04 '21 at 08:14
  • Yes I followed the steps there. I already looked at it before. I have cuda 10.2 11.4 and 10.0 installed. Should I delete the 10.2 and 10.0 version? – Thilo Rabens Jul 04 '21 at 11:15
  • It is recognizing my gpu, but isnt using it, when I run this: print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU'))) It prints out: Num GPUs Available: 1 – Thilo Rabens Jul 04 '21 at 11:38

0 Answers0