[Ubuntu] Lab Environment Setting

( Last update: 2020.01.17)

 

 

 Finally I have finished all the setups that I need right these days. 

 There have been seven steps.

 

 1. Installing Linux....Ubuntu in my Server PC by USB

 2. Setting my Office PC (Window installed) in my desk

 3. Installing some applications I use (Chrome, MicroOffice, KakaoTalk, MobaXterm ...otherwise)

 4. Connecting to my server by MobaXterm

 5. Installing Nvidia driver and CUDA, CUDNN

 6. Installing Anaconda in my server

   6.1. Setting up the environment

 7. Installing tensorflow-gpu and opencv in Anaconda env

   + Keras

 

 

 In this page, I'm gonna describe 4 steps in order.

---

 

1. Installing Ubuntu in Server PC by USB

   I borrowed a USB already containing Ubuntu file from my colleague. 

 Connect all the lines to my server PC and inserted USB there.

 Reboot pc with pressing ESC and Delete buttons at the same time. I cannot certain which button is right one, but anyway administor mod came up properly.

 

 

 

---

2. Setting my Office PC (Window installed) in my desk

   There were few problems in this step. Just connect all the lines that were connected with server PC before.

 I moved server PC to outside desk and set my office pc in my desk.

 

 

 

---

3. Installing some applications I use

(Chrome, MicroOffice, KakaoTalk, MobaXterm ...otherwise)

skip

 

 

 

---

4. Connecting to my server by MobaXterm

 1) Install MobaXterm free ver and start it.

 

 2) Press Session button

 

 

 3) Click SSH and fill in Basic SSH settings section.

 (Check 'Specify username)

 I wrote these datas at SSH article (private post)

 

 

 4) Success

 

 

---

5. Installing Nvidia driver and CUDA, CUDNN

1) Nvidia driver

 First, check whick one might be suitable for my environment.

 $ ubuntu-drivers devices

 Then, it shows me which one is the best (with 'recommended' letters)

 

 Install a driver.

 I got a reference to this below web site.

https://codechacha.com/ko/install-nvidia-driver-ubuntu/

우분투 18.04 - NVIDIA 드라이버를 설치하는 방법 | chacha

 

 

2) CUDA, CUDNN

 I finished installing CUDA in Ubuntu Environment with reference to this upper site.

https://hiseon.me/linux/ubuntu/cuda-install/

CUDA 설치 우분투 환경 - HiSEON

 

 

 

---

6. Installing Anaconda in my server

 

 I referred to this web site.

https://hiseon.me/python/ubuntu-anaconda-install/

우분투 아나콘다 설치 - HiSEON

 

 

 

---

7. Installing tensorflow-gpu and opencv in Anaconda env

 

 7-1) OpenCV

 

 

 7-2) Tensorflow-gpu

 

 (+)

+) Keras

 

 

 

 

 

---

 

 It looks like simple and may seem like there might be no any problems but it was difficult and complex.

 I don't want to do it again.