Use docker to configure python extension package pyhanlp on Linux server

1. Choose a suitable basic image, you can bring basic python3.6, C++ environment gcc, java SDK, etc., so you don’t need to install it in the container later

docker pull mirror command

docker pull [OPTIONS] NAME[:TAG|@DIGEST]

docker pull 镜像链接:版本号

2. Refer to the previous blog post to install various extension packages, including TensorFlow, jpype1, here you can use pip to install

Last blog post address: https://blog.csdn.net/qq_22472047/article/details/104474561

3. If java is not installed, you can download the java SDK installation package locally and upload it to the docker designated container

Download the specified version of the java SDK

Download link: https://www.oracle.com/cn/java/technologies/javase-jdk8-downloads.html

Create a folder where the java environment is stored

mkdir -p /usr/lib/jvm

Upload to the specified container of docker

1) Obtain the container id

docker ps -a

2) Get the complete container id

docker inspect -f  '{
   
   {.Id}}' container_id

3) Upload local files to the docker container

docker cp 本地文件路径 ID全称:容器路径

4) Unzip the java environment installation package

tar -zxvf jdk-8u131-linux-x64.tar.gz -C /usr/lib/jvm

5) Set environment variables

vim /etc/profile

When using a docker container, sometimes vim is not installed inside, and when you press the vim command, it prompts: vim: command not found. At this time, you need to install vim, but when you press the apt-get install vim command, it prompts:

Reading package lists... Done
Building dependency tree       
Reading state information... Done
E: Unable to locate package vim

At this time you need to knock:

apt-get update

The function of this command is to synchronize the indexes of the sources listed in /etc/apt/sources.list and /etc/apt/sources.list.d so that the latest software packages can be obtained.

Wait until the update is complete and then type the command

apt-get install vim

You can successfully install vim

At this time, you can configure the java environment variable and edit the configuration file

export JAVA_HOME=/usr/lib/jvm/jdk1.8.0_231
export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
export PATH=$PATH:$JAVA_HOME/bin

6) Reload the configuration file

source /etc/profile

This allows the configuration to take effect immediately without restarting

7) Test it

java -version

display

java version "1.8.0_131"
Java(TM) SE Runtime Environment (build 1.8.0_131-b11)
Java HotSpot(TM) 64-Bit Server VM (build 25.131-b11, mixed mode)

That is, the configuration is successful, and the whole installation process is over.

4. Install pyhanlp

pip install pyhanlp -i https://pypi.doubanio.com/simple/ --trusted-host pypi.doubanio.com

5. Check if pyhanlp is installed successfully

After entering the python environment, import the pyhanlp package, after success, no error will be reported and some necessary model files will be loaded

from pyhanlp import *