ニューマシンがいきなり起動しなくなったので、再セットアップする

この記事は，下書きでも良いくらい．
ニューマシン(())がいきなり起動しなくなったので，複数回のセットアップを想定して手順を記録します．

0. Ubuntu 18.04 LTS

Japanese ver., is a little inconvenient for the directory name and so on.
Thus, this time I installed English ver.
Note that you have to select Japanese keybourd.

1. NVIDIA Driver

• check device

lspci | grep -i nvidia

• check driver currently installed

dpkg -l | grep nvidia  
dpkg -l | grep cuda

• Remove if an old ver. is installed

sudo apt-get --purge remove nvidia*  
sudo apt-get --purge remove cuda*
sudo apt autoremove
sudo rm /usr/local/cuda
sudo rm -r /urs/local/cuda-*
reboot

• Install new driver

sudo add-apt-repository ppa:graphics-drivers/ppa  
sudo apt update  
sudo apt-get install nvidia-driver-440  
reboot

• check installation

nvidia-smi

2. CUDA

Note that you have to install CUDA ver. 10.0 for using tensorflow 1.14.0.
Ver. 10.1 is not compatible.

• check libraries currently installed

dpkg -l | grep cuda

• download from official HP
  https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1604&target_type=debnetwork

• install

sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub  
sudo dpkg -i cuda-repo-ubuntu1804_10.0.130-1_amd64.deb  
sudo apt update  
sudo apt install cuda-toolkit-10-0
reboot

• edit .bashrc

export PATH="/usr/local/cuda/bin:$PATH"
export LD_LIBRARY_PATH="/usr/local/cuda/lib64:$LD_LIBRARY_PATH"

• check installation

nvcc -V echo $PATH echo $LD_LIBRARY_PATH

3. cudnn

• download the following 3 files from official HP libcudnn7_7.4.2.24-1+cuda10.0_amd64.deb
  libcudnn7-dev_7.4.2.24-1+cuda10.0_amd64.deb
  libcudnn7-doc_7.4.2.24-1+cuda10.0_amd64.deb

• install

sudo dpkg -i libcudnn7_7.4.2.24-1+cuda10.0_amd64.deb
sudo dpkg -i libcudnn7-dev_7.4.2.24-1+cuda10.0_amd64.deb
sudo dpkg -i libcudnn7-doc_7.4.2.24-1+cuda10.0_amd64.deb

4. tensorflow

• install dependant libraries

sudo apt-get update  
sudo apt-get upgrade  
sudo apt-get install python3-pip python3-dev  
sudo apt-get install build-essential cmake git unzip pkg-config  
sudo apt-get install libopenblas-dev liblapack-dev  
sudo apt-get install libhdf5-serial-dev python-h5py  
sudo apt-get install python3-numpy  
sudo apt-get install python3-scipy  
sudo apt-get install python3-matplotlib  
sudo apt-get install ipython3  
sudo apt-get install graphviz  
sudo apt-get install python-opencv  
sudo pip3 install opencv-python  
sudo pip3 install pydot-ng  

• install tensorflow

sudo pip3 install tensorflow-gpu==1.14.0

• It is recommended to download the wheel for saving the environment https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.14.0-cp35-cp35m-linux_x86_64.whl

• install keras

pip3 install keras

• check if GPU works

$ python3  
from tensorflow.python.client import device_lib  
device_lib.list_local_devices()

• build cuda sample

mkdir ~/CUDA
cp -r /usr/local/cuda-10.0/samples ~/CUDA
cd ~/CUDA/samples
make
cd bin/x86_64/linux/release
./volumeRender

5. Japanese

click "manage installed languages"

click "install"

click "+"

click "Japanese"

selece "Japanese"MOZC"

6. Softwares

vs-code

• download deb package
  https://code.visualstudio.com/docs/setup/linux

• install

sudo apt-get install ./code_1.30.0-1544567151_amd64.deb

• check version

code --version

• install extensions
  markdown PDF
  Python

shutter

sudo add-apt-repository ppa:ubuntuhandbook1/shutter  
sudo apt update  
sudo apt install shutter

meld

• install using Ubuntu software

7. Libraries for NLP

Beautiful Soup, selenium

pip3 install beautifulsoup4  
pip3 install selenium

Dictionary(mecab-ipadic-NEologd)

• clone and install

git clone --depth 1 https://github.com/neologd/mecab-ipadic-neologd.git  
cd mecab-ipadic-neologd  
apt install curl  
./bin/install-mecab-ipadic-neologd -n -a

• edit /etc/mecabrc as:

dicdir = /usr/lib/mecab/dic/mecab-ipadic-neologd

mecab

sudo apt-get install mecab libmecab-dev mecab-ipadic mecab-ipadic-utf8  
pip3 install mecab-python3

wordcloud

pip3 install wordcloud

jumanpp(juman++)

• install boost

sudo apt install libboost-dev  
sudo apt install libboost-all-dev

• download and install

cd ~/Downloads
wget http://lotus.kuee.kyoto-u.ac.jp/nl-resource/jumanpp/jumanpp-1.02.tar.xz  
tar xJvf jumanpp-1.02.tar.xz  
cd jumanpp-1.02  
./configure  
make  
sudo make install  
rm -r jumanpp-1.02 (if necessary)

• check version

jumanpp -v

• test

echo "すもももももももものうち" | jumanpp

• install python interface

pip3 install pyknp