Ubuntu 18.04
RTX 2080 Super
nvidia driver 430 + Cuda 10.1 + cudnn 7.6
Tensorflow 2.1.0
1. Install Nvidia driver, cuda 10.1 and cudnn 7.6 with instruction from tensorflow website:
https://www.tensorflow.org/install/gpu#ubuntu_1804_cuda_101
2. Update bash profile
nano ~/.bashrc
$ NVIDIA CUDA Toolkit
Then insert:
export PATH=/usr/local/cuda-10.1/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-10.1/lib64
$ source ~/.bashrc
3. install tensorflow-gpu 2.1.0
4. Insert magic code in your code
# magic code for cudnn to sccussfuly launch
import tensorflow as tf
gpu_devices = tf.config.experimental.list_physical_devices('GPU')
for device in gpu_devices:
tf.config.experimental.set_memory_growth(device, True)
5. Use terminal to run your code, it doesn't work on VS code