How to install Tensorflow on ubuntu 16.04 server with GEForce RTX 2080
Bhaskar Chandra Trivedi
Last updated on
2 min read
- Install cuda 9.0 without nvidia driver.
- sudo chmod +x cuda_9.0.176_384.81_linux.run
- sudo ./cuda_9.0.176_384.81_linux.run - -extract=/home/username/Dirname
- sudo ./cuda-linux.9.0.176-22781540.run
- sudo ./cuda-samples.9.0.176-22781540-linux.run
- install patch 1
- download patch 1
- sudo sh cuda_9.0.176.1_linux.run
- download patch 2
- sudo sh cuda_9.0.176.2_linux.run
- download patch 3
- sudo sh cuda_9.0.176.3_linux.run
- download patch 4
- sudo sh cuda_9.0.176.4_linux.run
- set cuda path variable
- sudo cp ~/.bashrc ~/.bashrc_backup
- sudo nano ~/.bashrc
- export PATH=“/usr/local/cuda-9.0/bin:$PATH”
- export LD_LIBRARY_PATH=“/usr/local/cuda-9.0/lib64:$LD_LIBRARY_PATH”
- Install required cuDNN SDK for cuda 9.0.
- Download Anaconda and install Anaconda3-2018.12-Linux-x86_64.
- sudo bash Anaconda3-2018.12-Linux-x86_64.sh.
- sudo shutdown -r now.
- Create virtual env.
- conda create -n worldmodel python=3.5.4 numpy=1.13.3
- logout user(sometime required)
- source activate worldmodel
- install tensorflow.
- pip install –upgrade pip
- pip install –ignore-installed –upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.12.0-cp35-cp35m-linux_x86_64.whl
.
- possible error
- mkl-random 1.0.1 requires cython, which is not installed.
- Install cython
- pip install –upgrade pip
- pip install cython
- Verify Tensorflow installation.
- python
- import tensorflow
- If installtion fail import will throw some error msg.
