I am programming in Python 3.8 with Tensorflow installed along with my natural language processing project. When I want to begin the training phase, I get this message right before I begin...
2021-03-09 14:57:38.725406: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not
load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found
2021-03-09 14:57:38.725651: 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.
This DLL file should come preinstalled with the Tensorflow version I installed as later versions don't require the antiquated method of selecting Tensorflow CPU or GPU. I installed through Pip like this...
pip install tensorflow
I don't understand why this may happen. I have RTX 2070 in the system but I require this Library and DLL to train on the GPU. Can anyone explain why this may be?
UPDATE:
After I took advice from the answer below, I no longer received the error. However, it seems that it still uses the CPU over GPU for no reason. Here is what it shows when I begin training:
loading dataset...
Vocab size: 8332
Number of samples: 44098
2021-03-10 08:40:15.175461: 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.
creating model...
training model...
Step #0 Loss: 2.833855
Is this bad or is this just a setting to direct training towards the CPU?
