I want to run my R script on a GPU server. My script trains a regression model by Keras. to speed up the process, I used parallelization. Then I have done the below steps and then run my model:
First I installed a miniconda,
*Creating a Conda environment
*activated the conda invironment
*Installing PyTorch, pip, and tensorflow by (conda install tensorflow-gpu):
**conda install cudatoolkit=11.2 -c conda-forge
**pip install torch
**conda install tensorflow-gpu
**conda install cudatoolkit=11.1 pytorch=1.8.1 python=3.8 torchvision -c pytorch -c conda-forge
**pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.8.0+cu111.html
Then I ran my script.
The problem is, that GPU is not active, because of Tensorflow. So It is like a CPU and the running time is too long. These are the warning over running the script:
As far as I know I coudnt install tensorflow-gpu which is related to R!
#####---
2022-09-12 13:13:30.434379: 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; LD_LIBRARY_PATH: /usr/lib/R/lib:/usr/lib/x86_64-linux-gnu:/usr/lib/jvm/default-java/lib/server:/usr/lib/R/lib:/usr/lib/x86_64-linux-gnu:/usr/lib/jvm/default-java/lib/server
2022-09-12 13:13:30.434395: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] 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 GPU 支援 | TensorFlow for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2022-09-12 13:13:30.434750: 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 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
###-------
I need to mention the script did run but I couldn't improve the running time because I couldn't use the GPU because of the above warnings.
Is there anybody who knows how to fix the problem?
Should I add my code here?