Drawing library Matplotlib installation configuration

1. Matplotlib brief introduction

Matplotlib  yes  Python  An important part of the ecosystem , Is a drawing library for visualization , It offers a complete set and  matlab  Similar orders  API  And visual interface , It can generate exquisite graphics of publishing quality level ,Matplotlib  Make drawing very simple , Excellent balance between ease of use and performance .

2. Matplotlib install

Matplotlib  The dependency package and  Matplotlib  Itself in the standard  Python  Packages are available in the repository with  wheel  In the form of documents . therefore , have access to  pip  The package management tool makes it easy to  Matplotlib  Installed in the  MacOS、Windows、Linux  Wait for the system . The installation method is the same as that of many third-party libraries , install  Matplotlib  Just... On the command line , Execute the following command ：

pip install -U matplotlib

It should be noted that ： To install  Matplotlib  library , Must have been installed on your computer  Python. If you are using  Jupyter Notebook, because  Jupyter Notebook  With many dependent Libraries , Such as  Numpy、Pandas、Matplotlib、Scikit Learn etc. , You don't have to install these libraries separately .
After installation, you can use the following command to check , Confirm successful installation ：

pip list

3. Matplotlib To configure

After successful installation , That is to say, you can  Python  Import and use... Like other libraries in  Matplotlib, Configuration without more files , Usually we import it and use an alias  mpl：

import matplotlib as mpl

Matplotlib  The configuration of is more used to modify the default style of drawing graphics ,Matplotlib  The configuration file contains various default graphic configuration information , We can customize the drawing style by modifying these configuration information and global parameters , These parameters can change the size of the drawing 、 The color scheme 、 Font and a series of information .
It can be done in many ways  Matplotlib  Drawing configuration , This paper mainly introduces the following three configuration methods The configuration file To configure 、 adopt  rcParams['param_name']  Dynamic configuration and passing  matplotlib.rc()  Function configuration .

3.1 Configure through the configuration file

Configuration files can also be divided into several different levels , If we want to change the default configuration used by all graphics , You need to modify the global default configuration ; And if we need to use different configurations according to different tasks , Or different users use different configurations , You need to modify the local configuration file , To be able to use different graphic configurations in different users and tasks . According to the scope of the configuration file , It can be divided into three levels ： Global profile 、 User level profile and current task profile .
Files at three levels of different systems are located in different directories , You can use the following code , View the path to the profile ：

import matplotlib as mplimport os#  Global configuration directory print(mpl.__path__)#  Current user configuration directory print(mpl.get_configdir())#  Current task configuration directory , That is, the current code running Directory print(os.getcwd())

• Global profile  mpl-data\matplotlibrc, be located  Matplotlib  Installation directory of , For example, in  Window  Install it under  D:\Program Files\Python39\lib\site-packages\matplotlib  Under the table of contents , The full file name of the global configuration file is  D:\Program Files\Python39\lib\site-packages\matplotlib\mpl-data\matplotlibrc, By default , Drawings are drawn using this profile .
• User level profiles  .matplotlib\matplotlibrc, Located under the user directory , for example , The user directory is  C:\Users\Brainiac\, Then the corresponding configuration file is  C:\Users\Brainiac\.matplotlib\matplotlibc; If this file does not exist , It can also be created and modified according to the global configuration file and user requirements .
• Current task profile  matplotlibrc, That is, it is located under the code running Directory , It can be used to customize the code of the current task  Matplotlib  To configure , By default, this file does not exist , That is, the global or current user profile is used by default , We can create this file as needed , And configure as needed .

After introducing the location of the configuration file , By looking at the global configuration file , Observe the relevant graphic attributes that can be configured in the configuration file , The following is an example of a global configuration file ：

NOTE： As you can see, it is not recommended to modify the global configuration file directly , You can copy this file to the user and profile directory or the current task profile directory , And make changes as needed . The format of configuration file is generally   Property name : Property value , The line width is configured as follows  1.5：

lines.linewidth: 1.5

3.2 adopt rcParams[‘param_name’] To configure

And if we just want to simply modify the custom configuration in the current file , You can use the  rcParams['param_name']  Faster modifications . By using the following code , You can view the properties that can be customized ：

import matplotlib as mpl#  You can use the following three ways print(mpl.rc_params())print(mpl.rcParamsDefault)print(mpl.rcParams) 

The output result is similar to that in the configuration file , The format is also   Property name : Property value ：

...font.size: 10.0font.stretch: normalfont.style: normalfont.variant: normalfont.weight: normal...

Use  rcParams['param_name']  The way to modify the configuration is as follows , among  param_name  Represents the property name ：

import matplotlib as mpl#  Change the line width to 2mpl.rcParams['lines.linewidth'] = 2#  Change the line color to red mpl.rcParams['lines.color'] = 'r'

in application , The two most commonly used configurations include the display of Chinese and Chinese minus signs , If not configured , Chinese and Chinese minus signs are not supported by default ：

Configure in the following way ：

import matplotlib as mpl# According to Chinese mpl.rcParams['font.sans-serif'] = ['SimHei']# Show minus sign mpl.rcParams['axes.unicode_minus']=False

After configuration, the graphics can display Chinese and Chinese symbols normally ：

3.3 adopt matplotlib.rc() Function configuration

We can also use  matplotlib.rc()  Function to configure , How to use it is as follows ：

import matplotlib as mpl#  Modify lineweight mpl.rc('lines', linewidth=2, color='g')

among  rc  The first argument to the function is  group  Represents the group to which the attribute belongs , Scope used to qualify attributes , For example, in the above example, the line width  linewidth  Belong to line  lines  Used to limit the effect only in the line bar , And for the line width of the coordinate axis  linewidth  It doesn't work , If you want to change the line width of the drawing including the coordinate axis  linewidth  You need to use ：

import matplotlib as mpl#  Change the lineweight of the entire drawing mpl.rc('axes', linewidth=2)