Ubuntu tesorflow-gpu install

2019/01/03 업데이트

Version check

  1. ubuntu 16.04
  2. nvidia driver setup

  3. cuda setup – 9.0

  4. cudnn setup -7.0

nvidia-smi check

nvidia-smi

CUDA version check

cat /usr/local/cuda/version.txt

CUDA VERSION 9.0.176

CUDNN version check

cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2

CUDNN_MAJOR 7

vultr-env 환경 셋팅

virtualenv –system-site-packages -p python3 ./tensorflow_py3

Tensorflow-gpu 설치

source tensorflow_py3/bin/activate

pip install –upgrade tensorflow-gpu

Tensorflow-gpu 사용 체크

import tensorflow as tf

sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))


 

에러 발생

  • cuda 9.0, cudnn 7.0로 해야 에러가 나지 않는다.

https://www.tensorflow.org/install/pip

How to set up Tensorflow with GPU CUDA 8 in virtualenv and Python 3.5

https://www.tensorflow.org/install/gpu

https://stackoverflow.com/questions/50622525/which-tensorflow-and-cuda-version-combinations-are-compatible

https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html

CUDA 설치 우분투 환경

https://keras.io/ – 케라스 설치

Keras Tutorial: How to get started with Keras, Deep Learning, and Python

Keras Tutorial: The Ultimate Beginner’s Guide to Deep Learning in Python

http://solarisailab.com/archives/303

https://tensorflowkorea.gitbooks.io/tensorflow-kr/content/g3doc/get_started/os_setup.html

http://hunkim.github.io/ml/

댓글 남기기

이메일은 공개되지 않습니다. 필수 입력창은 * 로 표시되어 있습니다