I have been trying to set up a virtual environment on the HPC Cluster, but hitting a wall everytime when it comes to tensorflow using GPU.
module load python3.8/3.8 eval "$(conda shell.bash hook)" conda activate graphgan
After this I use the following commands to set up the environment
conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
When i try to import tensorflow as tf it shows the following error
2024-01-26 23:06:17.968937: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (on eDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA To enable them in other operations, rebuild TensorFlow with the appropriate comp iler flags. 2024-01-26 23:06:18.289070: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cu BLAS when one has already been registered 2024-01-26 23:06:20.132284: W tensorflow/stream_executor/platform/default/dso_lo ader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfe r.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PA TH: /opt/ohpc/pub/apps/anaconda3/envs/py38/lib:/home/vanshg.phy21.iitbhu/.conda/ envs/graphgan/lib/ 2024-01-26 23:06:20.132710: W tensorflow/stream_executor/platform/default/dso_lo ader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: li bnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /opt/ohpc/pub/apps/anaconda3/envs/py38/lib:/home/vanshg.phy21. iitbhu/.conda/envs/graphgan/lib/ 2024-01-26 23:06:20.132755: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc: 38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned a bove are installed properly.
I know this will run my code without any issues but i want to use GPU acceleration. Could anyone please tell me what should I do to solve this
I was expecting for the GPU to get detected, as it did in the tutorial I saw. Instead it started throwing errors. I also checked in the directory it mentions and indeed the files were missing. What are the extra libraries that i have to install to make this work?