Conscience summary! Jupyter notebook from Xiaobai to master, the nanny tutorial is coming!

from ： Youer cottage

Hello everyone , I'm a rookie ！

Today, I'd like to share an article with you About data science tools Jupyter Notebook Of Quick start and tutorial , You can compare and learn ：

One 、 What is? Jupyter Notebook？

1. brief introduction

Jupyter Notebook Is a web based application for interactive computing . It can be applied to the whole process of calculation ： Development 、 Documentation 、 Run the code and display the results .——Jupyter Notebook The official introduction

In short ,Jupyter Notebook It's in the form of a web page , It can be in the web page 「 direct 」 Write and run code , The running results of the code will also be displayed directly under the code block . Such as in the process of programming need to write documentation , It can be written directly in the same page , Convenient for timely explanation and explanation .

2. Part of the

① Web applications

Web application is based on web form 、 Combined with the preparation of documentation 、 The mathematical formula 、 Interactive computing and other rich media tools .「 in short , Web application is a tool that can realize various functions .」

② file

namely Jupyter Notebook All interactive computing in 、 Document the instructions 、 The mathematical formula 、 Images and other rich media forms of input and output , It's all in the form of documents .

These documents are saved with the suffix .ipynb Of JSON Format file , Not only is it easy to version control , It's also easy to share .

Besides , Documents can also be exported as ：HTML、LaTeX、PDF Equiform .

3. Jupyter Notebook The main characteristics of

1. Programming with 「 Syntax highlighting 」、 Indent 、tab The function of complement .

2. You can run the code directly through the browser , At the same time, the running result is displayed under the code block .

3. Display the calculation results in rich media format . Rich media formats include ：HTML,LaTeX,PNG,SVG etc. .

4. When writing instruction documents or statements to code , Support Markdown grammar .

5. Support use LaTeX Write a mathematical note .

Two 、 install Jupyter Notebook

0. Try it first , Decide again

If you read the above right Jupyter Notebook You still can't decide whether your introduction is suitable for you , Then don't worry , You can start with 「 Installation free trial experience 」 once , Then make a decision .

It is worth noting that , The official simultaneous trial is limited , If you click the link, the page you enter is shown in the figure below , Then don't worry , Try again later .

Trial failed

If you are lucky enough , Then you will see the following interface , You can start to experience .

main interface

Edit page

1. install

① Installation premise

install Jupyter Notebook The premise is to install Python（3.3 Version and above , or 2.7 edition ）.

② Use Anaconda install

If you are little white , Then I suggest you install Anaconda To solve Jupyter Notebook Installation problems , because Anaconda Has been automatically installed for you Jupter Notebook And other tools , also python In more than 180 Science packs and their dependencies .

You can go through Anaconda Choose to download from the official download page of ;

Routinely , Installed Anaconda It has been automatically installed for you when you release Jupyter Notebook Of , But if it's not installed automatically , Then at the terminal （Linux or macOS Of “ terminal ”,Windows Of “Anaconda Prompt”, Hereinafter referred to as “ terminal ”） Enter the following command to install ：

conda install jupyter notebook

③ Use pip Command to install

If you are experienced Python The player , Want to try pip Command to install Jupyter Notebook, Then please take a look at the following steps ！ The next commands are input in the terminal ！

hold pip Upgrade to the latest version

pip install --upgrade pip

• Be careful ： The old version of pip In the installation Jupyter Notebook In the process or facing the problem that dependencies cannot be installed synchronously . therefore 「 Strongly recommended 」 The first pip Upgrade to the latest version .

install Jupyter Notebook

pip install jupyter

3、 ... and 、 function Jupyter Notebook

0. help

If you have any jupyter notebook Command questions , Consider checking the official help documentation , The order is as follows ：

jupyter notebook --help

or

jupyter notebook -h

1. start-up

① The default port starts

Enter the following command in the terminal ：

jupyter notebook

After executing the command , A series of messages will be displayed in the terminal notebook Server information , At the same time, the browser will start automatically Jupyter Notebook.

The contents displayed on the terminal during startup are as follows ：

$ jupyter notebook
[I 08:58:24.417 NotebookApp] Serving notebooks from local directory: /Users/catherine
[I 08:58:24.417 NotebookApp] 0 active kernels
[I 08:58:24.417 NotebookApp] The Jupyter Notebook is running at: http://localhost:8888/
[I 08:58:24.417 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).

• Be careful ： After the Jupyter Notebook All operations , Please keep the terminal 「 Do not shut down. 」, Because once the terminal is turned off , It will disconnect the link with the local server , You will not be able to Jupyter Notebook Other operations in .

By default, the browser address bar will display ：http://localhost:8888. among ,“localhost” It means the machine ,“8888” It's the port number .

If you 「 meanwhile 」 Several Jupyter Notebook, Due to the default port “8888” occupied , Therefore, the number in the address bar will start from “8888” rise , Start one more Jupyter Notebook The number is added 1, Such as “8889”、“8890”……

② Specify the port to start

If you want to customize the port number to start Jupyter Notebook, You can enter the following commands in the terminal ：

jupyter notebook --port <port_number>

among ,“<port_number>” Is a custom port number , Write it directly in the form of numbers in the command , No angle brackets around numbers “<>”. Such as ：jupyter notebook --port 9999, That is, when the port number is “9999” The server starts Jupyter Notebook.

③ Start the server without opening the browser

If you just want to start Jupyter Notebook But I'm not going to go to the main page right away , Then you don't need to start the browser immediately . Input in the terminal ：

jupyter notebook --no-browser

here , The started server information will be displayed on the terminal , And after the server starts , Show the link to open the browser page . When you need to launch a browser page , Just copy the link , And paste it in the address bar of the browser , Tap enter to change to your Jupyter Notebook page .

no_browser

In the example figure, I started multiple at the same time when I finished the above content Jupyter Notebook, So show me “8888” Port number occupied , Finally assigned to me is “8889”.

2. The main page

① Main page content

After executing the start command , The browser will go to Notebook Main page , As shown in the figure below .

If the folder in your homepage is different from mine , Or you're wondering why there are so many folders inside the first startup , Don't worry about , The folders here are all directory files in your home directory . You can execute the following in the terminal 2 Step to see ：

① cd or cd - or cd ~ or cd /Users/

• This command will enter your home directory .

• “<user_name>” Is the user name . No angle brackets around the user name “<>”.

② ls

• This command will display the files in your home directory .

② Set up Jupyter Notebook File storage location

If you don't want to put the future in Jupyter Notebook All documents written in are saved directly in the home directory , Then you need to modify Jupyter Notebook File storage path .

⑴ Create folder / Catalog

• Windows The user is trying to store Jupyter Notebook Of documents 「 disk 」 in 「 New folder 」 And name the folder ; Double click to enter the folder , Then copy the path in the address bar .

• Linux/macOS The user is trying to store Jupyter Notebook The location of the file 「 Create directory 」 And name the directory , The order is ：mkdir; Entry directory , The order is ：cd; View the path to the directory , The order is ：pwd; Copy the path .

• Be careful ：“<directory_name>” The custom directory name is . No angle brackets around the directory name “<>”.

⑵ Profile path

• A convenient command to get the path of the configuration file ：

jupyter notebook --generate-config

• Be careful ： Although this command can be used to view the path of the configuration file , But the main purpose is whether to set the configuration file under this path 「 Replace 」 by 「 Default profile 」. If this is your first inquiry , that 「 Maybe 」 The following prompt will not appear ; If the file already exists or has been modified , After using this command, you will be asked “Overwrite /Users/raxxie/.jupyter/jupyter_notebook_config.py with default config? [y/N]”, namely “ Overwrite the files in this path with the default configuration file ？”, If the “y”, Then complete the coverage , Then all previous modifications will be invalidated ; If it's just to query the path , Then be sure to enter “N”.

Under normal circumstances ,Windows and Linux/macOS The path where the configuration file is located and the name of the configuration file are as follows ：

• Windows System configuration file path ：C:\Users\\.jupyter\

• Linux/macOS System configuration file path ：/Users//.jupyter/ or ~/.jupyter/

• Profile name ：jupyter_notebook_config.py

• Be careful ：

① “<user_name>” For your user name . No angle brackets around the user name “<>”.

② Windows and Linux/macOS The storage path of the system configuration file is actually the same , It's just that the system is different , It's just in different forms .

③ Windows and Linux/macOS The configuration file of the system is the same . File name to “.py” ending , yes Python The executable of .

④ If you don't go to the location of the configuration file in one step , Instead, go into the folder layer by layer / The directory , So when you enter the home directory , use ls The command will find that “.jupyter” Folder / Catalog . This is because those who use “.” The first directories are hidden files , You can go through ls -a Command to view all hidden files in the current location .

⑶ Modify the configuration file

• Windows Users of the system can use document editing tools or IDE open “jupyter_notebook_config.py” File and edit . Common document editing tools and IDE There is a notepad 、Notepad++、vim、Sublime Text、PyCharm etc. . among ,vim There is no graphic interface , Is a steep learning curve editor , Other tools are not described here , Because it's relatively easy to get started . adopt vim To modify the configuration file, please continue to read .

• Linux/macOS Users of the system suggest calling directly through the terminal vim To modify the configuration file . The specific operation steps are as follows ：

⒜ Open profile

Open the terminal , Enter the command ：

vim ~/.jupyter/jupyter_notebook_config.py

vim Open profile

After executing the above command, you will enter the configuration file .

⒝ Search for keywords

Search for keywords after entering the configuration file “c.NotebookApp.notebook_dir”. The search method is as follows ：

Do not press other keys after entering the configuration file , use 「 Half width in English 」 Direct input /c.NotebookApp.notebook_dir, At this time, the search keywords have been highlighted in the document , Press enter , The cursor switches from the bottom to the first letter of the searched keyword in the body of the document .

⒞ Edit profile

Press 「 A lowercase letter i」 Enter edit mode , The bottom appears “--INSERT--” It indicates that you have successfully entered editing mode . Use the arrow keys to position the cursor on the second single quotation mark （ Which character is the cursor positioned on , Start typing just before this character ）, hold “⑴ Create folder / Catalog ” Paste the path copied in step here .

⒟ uncomment

Put the beginning of the line ** Well No （#）** Delete . Because the profile is Python The executable of , stay Python in , Well No （#） Notation , That is, the command line will not be executed during compilation , So in order for the changes to take effect , You need to delete the pound sign （#）.

config

⒠ Save configuration file

According to the first ESC key , Exit from edit mode , Go back to command mode .

Reuse 「 Half width in English 」 Direct input :wq, Press enter to save and exit the configuration file successfully .

Be careful ：

• 「 The colon （:）」 Be sure to have , And also 「 Half width in English 」.

• w： preservation .

• q： sign out .

⒡ verification

Enter command in terminal jupyter notebook open Jupyter Notebook, At this point, you will see a refreshing interface , Congratulations ！

modified

⒢ Be careful

• All the above commands are in 「 Half width in English 」 Format input , If there is an error , Please check these two conditions strictly ,「 english 」 And 「 Half angle 」.

• Here we only introduce vim How the editor modifies the configuration file , No, right vim Explain the detailed use of the editor , So there's no need to know vim How to use the editor , Just follow the above steps and you will be able to successfully complete the modification ！

• It is recommended to learn when you have time and experience vim Use of editor . This powerful editor will become a sharp weapon in your future work .

Four 、Jupyter Notebook Basic use of

1. Files page

Files page

At this time, there should be no... In your interface “Conda” and “Nbextensions” category . Don't worry , These two categories will be in “ 5、 ... and 、 Expand functions ” Medium “1. relation Jupyter Notebook and conda Environment and package ——‘nb_conda’” and “2.Markdown Generate Directory ” Install in .

Files Pages are used to manage and create categories related to files .

For existing files , You can check the file , Copy the selected file 、 rename 、 Move 、 download 、 see 、 Edit and delete operations .

meanwhile , Or as needed , stay “New” Select the environment where you want to create the file from the drop-down list , Create “ipynb” Format notebook 、“txt” The format of the document 、 Terminal or folder . If the environment you created does not appear in the drop-down list , Then you need to go to “ 5、 ... and 、 Expand functions ” Medium “1. relation Jupyter Notebook and conda Environment and package ——‘nb_conda’” and “ 6、 ... and 、 Add the kernel ——‘ipykernel’” Solve this problem in .

① Basic operation of notebook

Use of notebook

The figure above shows the basic structure and functions of the notebook . According to the annotations in the figure, most of the usage problems can be solved ！

The use of the toolbar is as intuitive as the annotation in the figure , There is no more explanation here . One thing to note is that “ The state of the cell ”, Yes Code,Markdown,Heading,Raw NBconvert. among , The most commonly used are the first two , Code status ,Markdown Write status .Jupyter Notebook It has been cancelled Heading state , The Title Cell . In its place Markdown From level one to level six . and Raw NBconvert It is rarely used at present , I won't explain too much here .

The menu bar covers all the functions of the notebook , Even if it's a toolbar , You can also find it in the category of the menu bar . However , Not all functions are commonly used , such as Widgets,Navigate.Kernel Use of categories , Mainly the operation of the kernel , Like interruptions 、 restart 、 Connect 、 close 、 Switch kernel, etc , Because we have selected the kernel when creating the notebook , Therefore, the operation of switching the kernel is convenient for us to switch to the kernel environment we want when using the notebook . Due to relatively conventional functions , It's very convenient to try to use the function of notebook according to the notes in the figure , So I won't explain it in detail .

② There are two ways to rename a notebook

⑴   Rename inside the notebook

When using a notebook , You can rename it directly inside . On the top left “Jupyter” Next to the icon is the default title of the program “Untitled”, Click on “Untitled” Then enter the self drawn title in the pop-up dialog box , Click on “Rename” The Rename is complete .

⑵   Rename the outside of the notebook

If you forget to rename your notebook when using it , And has been saved and exited to “Files” Interface , It's in “Files” Check the file to be renamed in the interface , Click on “Rename” Then directly enter the self-made title .

⑶ demonstration

rename

2. Running page

Running The page mainly shows the currently running terminals and “ipynb” Format notebook . If you want to close the open terminal and “ipynb” Format notebook , You can't exit the program completely just by closing its page , Need to be in Running Click its corresponding... On the page “Shutdown”. More closing methods can be found in “ 8、 ... and 、 Close and exit ” Medium “1. Turn off the laptop and terminal ”

Running

3. Clusters page

Clusters tab is now provided by IPython parallel. See 'IPython parallel' for installation details.

*

Clusters The category has now been changed from IPython parallel docking , And due to the low frequency of use at this stage , Therefore, no detailed description will be made here , For more information, visit IPython parallel Official website .

4. Conda page

Conda The page is mainly Jupyter Notebook And Conda After the association, the Conda A page tool for direct operation and management of environment and package . For details, please refer to “ 5、 ... and 、 Expand functions ” Medium “1. relation Jupyter Notebook and conda Environment and package ——‘nb_conda’”. This is currently used Jupyter Notebook A necessary part of , So be sure to check .

5. Nbextensions page

nbextensions

Nbextensions The page provides multiple Jupyter Notebook Plug in for , Make it more powerful . The main plug-ins used in this page are nb_conda,nb_present,Table of Contents(2). We don't need to fully master these functions , You don't need to install all the extensions , According to the learning ideas provided in this document , We just need to install Talbe of Contents(2) that will do , This function can be Markdown Documents provide directory navigation , It is convenient for us to write documents . Please refer to the installation instructions “ 5、 ... and 、 Expand functions ” Medium “2.Markdown Generate Directory ”.

5、 ... and 、 Expand functions

1. relation Jupyter Notebook and conda Environment and package ——“nb_conda”*

① install

conda install nb_conda

The execution of the above command can make you conda Create an environment with Jupyter Notebook Related to , It's convenient for you to Jupyter Notebook In use , Create notebooks to work in different environments .

② Use

• Can be in Conda Under the category of conda Environment and package for a series of operations .

  
  conda category

• It can be in the notebook “Kernel” In the category “Change kernel” Switch kernel .

  
  Switch kernel

③ uninstall

canda remove nb_conda

Execute the above command to uninstall nb_conda package .

2. Markdown Generate Directory

• Different from Youdao cloud notes Markdown compiler ,Jupyter Notebook Unable to Markdown Documents add directories through specific syntax , Therefore, you need to install extensions to add directories .

conda install -c conda-forge jupyter_contrib_nbextensions

• After execution of the above order , start-up Jupyter Notebook, You will find that there are too many navigation bars “Nbextensions” Category of , Click on “Nbextensions”, Check “Table of Contents ⑵”

  
  nbextensions

• After that Jupyter Notebook Use in Markdown, Click the icon below to use .

  
  Add a directory

3. Markdown Set up links in the text and locate

In the use of Markdown When editing a document , It is inevitable to encounter the need to set links in the text , Locate elsewhere in the document for easy viewing . because Markdown Can be perfectly compatible html grammar , Therefore, this function can be realized through html In grammar “a label ” Index usage to achieve .

The syntax is as follows ：

[ Add the body of the link ](# Custom index words )
<a id= Custom index words > Jump tips </a>

• Be careful ：

1. All symbols in the grammatical format are 「 Half width in English 」.

2. “ Custom index words ” Best in English , Longer words can be underlined .

3. “a label ” Appear in the position of the article you want to jump to ,html Except for single label, all labels shall comply with “ Have a head （） There must be a tail （）” Principles . Between the head and the tail “ Jump tips ” It's optional .

4. “a label ” Medium “id” The value is set when adding a link to the body “ Custom index value ”, Through here “id” The value of realizes the function of jumping from the link of the text to the specified position .

example ：

1. There are jump hints

2. No jump prompt

4. Load the specified web page source code

① Use scenarios

Want to be in Jupyter Notebook Directly load the source code of the specified website into the notebook .

② Method

Execute the following command :

%load URL

among ,URL Specify the address of the web site for .

③ example

Load the source code of the specified website

5. Load local Python file

① Use scenarios

Want to be in Jupyter Notebook Load local Python File and execute the file code .

② Method

Execute the following command ：

%load Python The absolute path to the file 

③ Be careful

1. Python The suffix of the file is “.py”.

2. “%load” In the heel Python Of documents 「 Absolute path 」.

3. After typing the command , Can press CTRL enter To execute an order . First execution , It's going to be local Python Load the contents of the file into the cell . here ,Jupyter Notebook Will automatically “%load” Command comment out （ That is, add a well number in front of “#”）, So that the command is not repeated when executing the loaded file code ; Second execution , Is the code that executes the loaded file .

④ example

Load local Python file

6. Run the local server directly Python file

① Use scenarios

Do not want to be in Jupyter Notebook Load local data in the cell of Python file , Want to run directly .

② Method

Carry out orders ：

%run Python The absolute path to the file 

or

!python3 Python The absolute path to the file 

or

!python Python The absolute path to the file 

③ Be careful

1. Python The suffix of the file is “.py”.

2. “%run” In the heel Python Of documents 「 Absolute path 」.

3. “!python3” Used to perform Python 3.x Version code .

4. “!python” Used to perform Python 2.x Version code .

5. “!python3” and “!python” Belong to !shell command The use of grammar , That is to say Jupyter Notebook In the implementation of shell The syntax of the command .

6. After typing the command , Can press CTRL enter To execute an order , Local... Will not be displayed during execution Python The content of the document , Directly display the operation results .

④ example

Direct operation Python file

7. stay Jupyter Notebook Get the current location in

① Use scenarios

Want to be in Jupyter Notebook Get the current location in 「 Absolute path 」.

② Method

%pwd

or

!pwd

③ Be careful

1. The location obtained is the current location Jupyter Notebook The location of the notebook created in , And the position is 「 Absolute path 」.

2. “!pwd” Belong to !shell command The use of grammar , That is to say Jupyter Notebook In the implementation of shell The syntax of the command .

④ example

Get the current path

8. stay Jupyter Notebook Use shell command

① Method 1 —— In the cell of the notebook

⑴ grammar

!shell command 

• stay Jupyter Notebook Use English exclamation marks in the notebook cells in “!” Followed by shell Orders can be executed shell command .

⑵ example

shell command

② Method 2 —— stay Jupyter Notebook New terminal in

⑴ Starting method

stay Jupyter Notebook main interface , namely “File” Click... In the interface “New”; stay “New” Click in the drop-down box “Terminal” That is, a new terminal . At this time, the terminal location is in your home directory , Can pass pwd Command to query the absolute path of the current location .

⑵ Close method

stay Jupyter Notebook Of “Running” In the interface “Terminals” You can see the running Terminal in the category , Click on the following “Shutdown” You can turn off the terminal .

⑶ example

terminal

9. Hide notebook input cells

① Use scenarios

stay Jupyter Notebook In your notebook, whether writing documents or programming , All have input （In []） And the output （Out []）. When we write code or documents that use more cells , Sometimes we just want to focus on the output content and don't look at the input content for the time being , In this case, you need to hide the input cells and only show the output cells .

② Method 1

⑴ Code

from IPython.display import display
from IPython.display import HTML
import IPython.core.display as di # Example: di.display_html('<h3>%s:</h3>' % str, raw=True)

#  The purpose of this line of code is ： When a document is used as HTML Format output , The input cells will be hidden by default .
di.display_html('<script>jQuery(function() {if (jQuery("body.notebook_app").length == 0) { jQuery(".input_area").toggle(); jQuery(".prompt").toggle();}});</script>', raw=True)

#  This line of code will be added “Toggle code” Button to switch “ hide / Show ” Enter cell .
di.display_html('''<button onclick="jQuery('.input_area').toggle(); jQuery('.prompt').toggle();">Toggle code</button>''', raw=True)

Enter the above code in the first cell of the notebook , And then execute , You can use... In this document “ hide / Show ” Enter cell function .

• defects ： This method is not very suitable for Markdown Cell .

⑵ example

hide / Display method 1

③ Method 2

⑴ Code

from IPython.display import HTML

HTML('''<script>
code_show=true; 
function code_toggle() {
 if (code_show){
 $('div.input').hide();
 } else {
 $('div.input').show();
 }
 code_show = !code_show
} 
$( document ).ready(code_toggle);
</script>
<form action="javascript:code_toggle()"><input type="submit" value="Click here to toggle on/off the raw code."></form>''')

Enter the above code in the first cell of the notebook , And then execute , You can use... In this document “ hide / Show ” Enter cell function .

• defects ： This method is not very suitable for Markdown Cell .

⑵ example

hide / Display method 2

10. Magic command

At present, there are not too many magic commands , Therefore, for the time being, please refer to the documents on the official website .

6、 ... and 、 Add the kernel ——“ipykernel” *

1. Use scenarios

1. Scene one ： Use different versions of Python Work on , stay Jupyter Notebook Cannot switch in , namely “New” The required environment cannot be used in the drop-down menu of .

2. Scene two ： Created different virtual environments （ Maybe have the same Python Version, but the package installed is different ）, stay Jupyter Notebook Cannot switch in , namely “New” The required environment cannot be used in the drop-down menu of .

Next, we will use “ Command line mode ” and “ GUI mode ” To solve the problems of the above two scenarios . seeing the name of a thing one thinks of its function ,“ Command line mode ” That is to solve the problem step by step by executing commands in the terminal ;“ GUI mode ” It's through Jupyter Notebook Solve the above problems by clicking the mouse on the web page .

“ Command line mode ” It seems complicated , And it also divides the use scenarios , But solving the problem in this way can better understand the working principle , such as , What is the command corresponding to each step of operation , And what purpose is the execution of the command to achieve , These may be completed by a click action encapsulated in the graphical interface . For friends who want to know more about its operation process , You can read down .

2. The command line mode of the solution

① Use different versions of Python

⑴ stay Python 3 Created in Python 2 kernel

⒜ pip install

• First installation Python 2 Of ipykernel package .

python2 -m pip install ipykernel

• Again 「 The current user 」 install Python 2 The kernel of （ipykernel）.

python2 -m ipykernel install --user

• Be careful ：“--user” Parameter means that it is installed for the current user , Instead of system wide installation .

⒝ conda install

• First create Python Version is 2.x And ipykernel The new environment , among “<env_name>” Name the custom environment , No angle brackets around the environment name “<>”.

conda create -n <env_name> python=2 ipykernel

• Then switch to the newly created environment .

Windows: activate <env_name>
Linux/macOS: source activate <env_name>

• by 「 The current user 」 install Python 2 The kernel of （ipykernel）.

python2 -m ipykernel install --user

• Be careful ：“--user” Parameter means that it is installed for the current user , Instead of system wide installation .

⑵ stay Python 2 Created in Python 3 kernel

⒜ pip install

• First installation Python 3 Of ipykernel package .

python3 -m pip install ipykernel

• Again 「 The current user 」 install Python 2 The kernel of （ipykernel）.

python3 -m ipykernel install --user

• Be careful ：“--user” Parameter means that it is installed for the current user , Instead of system wide installation .

⒝ conda install

• First create Python Version is 3.x And ipykernel The new environment , among “<env_name>” Name the custom environment , No angle brackets around the environment name “<>”.

conda create -n <env_name> python=3 ipykernel

• Then switch to the newly created environment .

Windows: activate <env_name>
Linux/macOS: source activate <env_name>

• by 「 The current user 」 install Python 3 The kernel of （ipykernel）.

python3 -m ipykernel install --user

• Be careful ：“--user” Parameter means that it is installed for the current user , Instead of system wide installation .

② Create kernels for different environments

⑴ Switch to the environment where the kernel needs to be installed

Windows: activate <env_name>
Linux/macOS: source activate <env_name>

• Be careful ：“<env_name>” Is the name of the environment where the kernel needs to be installed , No angle brackets around the environment name “<>”.

⑵ Check whether the environment has ipykernel package

conda list

Execute the above command to view the packages installed in the current environment , If not installed ipykernel package , Then execute the installation command ; Otherwise proceed to the next step .

conda install ipykernel

⑶ Install... For the current user in the current environment Python kernel

• If the environment Python Version is 2.x, Then execute the command ：

python2 -m ipykernel install --user --name <env_name> --display-name "<notebook_name>"

• If the environment Python Version is 3.x, Then execute the command ：

python3 -m ipykernel install --user --name <env_name> --display-name "<notebook_name>"

• Be careful :

1. “<env_name>” Is the name of the current environment . No angle brackets around the environment name “<>”.

2. “<notebook_name>” For custom display in Jupyter Notebook The name of the . There are no angle brackets around the name “<>”, but 「 Double quotation marks must be added 」.

3. “--name” The value of the parameter , namely “<env_name>” yes Jupyter For internal use , The storage path of its directory is ~/Library/Jupyter/kernels/. If the defined name already exists in the path , Then the contents of the directory with this name will be automatically overwritten .

4. “--display-name” The value of the parameter is displayed in Jupyter Notebook The name in the menu .

⑷ test

Use command jupyter notebook start-up Jupyter Notebook; stay “Files” Under the “New” The drop-down box will find you in the ⑶ Custom name in step , here , You can enjoy yourself in Jupyter Notebook Switch environment in , Create notebooks to work and study in different environments ！

3. Graphical interface mode of solution

① You created a new environment , But I found it in Jupyter Notebook Of “New” This environment cannot be found in , Cannot create notebook in this environment .

The graphical interface _ problem

② Get into Jupyter Notebook → Conda → stay “Conda environment” Click you want to add ipykernel Package environment → Enter... In the search box at the bottom left “ipykernel” → Check “ipykernel” → Click... Next to the search box “→” arrow → Installation completed → Find... In the lower right box “ipykernel” The installation is successful .

The graphical interface _ solve

③ At terminal CTRL C close Jupyter Notebook Then restart your server Jupyter Notebook, stay “File” Of “New” You can find your environment in the drop-down list of .

The graphical interface _ verification

7、 ... and 、Jupyter Notebook Shortcut key

1. Mac And Windows Special key comparison table

Symbol	Mac Key	Windows Key
⌘	command	nothing
⌃	control	ctrl
⌥	option	alt
⇧	shift	shift
	return	return
␣	space	space
⇥	tab	tab
⌫	delete	backspace
⌦	fn  delete	delete
-	-	-

2. Jupyter Notebook Two modes of notebook

① Command mode

• Command mode combines keyboard commands with Jupyter Notebook Combination of notebook commands , You can run notebook commands through the combination of different keys on the keyboard .

• Press esc Key to enter command mode .

• In command mode , The cell border is gray , And the left border line is blue thick line .

Command mode

② Edit mode

• Edit mode allows you to edit code or documents within cells .

• Press enter or return Key to enter edit mode .

• In edit mode , Both the cell border and the left border line are green .

Edit mode

3. Shortcut keys for two modes

① Command mode

Shortcut key	purpose
F	Find and replace
	Enter edit mode
⌘⇧F	Open the command palette
⌘⇧P	Open the command palette
P	Open the command palette
⇧	Run the current cell and select the next cell
⌃	Run the selected cell
⌥	Run the current cell and insert the cell below
Y	Switch the cell to code state
M	Switch the cell to markdown state
R	Switch the cell to raw state
1	Set the cell to the first level title
2	Set the cell as a secondary title
3	Set the cell to a three-level title
4	Set the cell to a four level title
5	Set the cell to a five level title
6	Set the cell to a six level title
↑	Select the upper cell
K	Select the upper cell
↓	Select the lower cell
J	Select the lower cell
⇧K	Select more cells up
⇧↑	Select more cells up
⇧J	Select multiple cells down
⇧↓	Select multiple cells down
A	Insert cells above
B	Insert cells below
X	Cut the selected cell
C	Copy selected cells
⇧V	Paste into the upper cell
V	Paste into the lower cell
Z	Undo delete
D, D	Delete selected cells
⇧M	Merge selected cells , If you select one directly, it will be merged with the next cell
⌘S	preservation
S	preservation
L	Convert line number
O	Conversion output
⇧O	Convert scrolling output
H	Show shortcut help
I, I	interrupt Notebook kernel
O, O	restart Notebook kernel
esc	Close page
Q	Close page
⇧L	Convert all cell line numbers and the settings remain valid
⇧␣	Scroll up
␣	Scroll down

② Edit mode

Mac Shortcut key	Windows Shortcut key	purpose
⇥	⇥	Code completion or indentation
⇧⇥	⇧⇥	Tips
⌘]	⌃]	To indent backward
⌘[	⌃[	To indent forward
⌘A	⌃A	Future generations
⌘Z	⌃Z	revoke
⌘/		notes
⌘D		Delete this line
⌘U		revoke
⌘↑	⌃↑	The cursor jumps to the starting position of the cell
⌘↓	⌃↓	The cursor jumps to the final position of the cell
⌥←	⌃←	Move the cursor one word to the left
⌥→	⌃→	Move the cursor one word to the right
⌥⌫	⌃⌫	Delete the previous word
⌥⌦	⌃⌦	Delete the next word
⌘⇧Z	⌃Y	redo
⌘⇧U	⌃⇧Z	redo
⌘⌫	⌃⌫	Delete the content to the left of the cursor in this line
⌘⌦	⌃⌦	Delete the content to the right of the cursor in this line
⌃M	⌃M	Enter command mode
esc	esc	Enter command mode
⌘⇧F		Open the command palette
⌘⇧P		Open the command palette
⇧	⇧	Run the current cell and select the next cell
⌃	⌃	Run the selected cell
⌥	⌥	Run the current cell and insert the cell below
⌃⇧-	⌃⇧-	Split the cell at the cursor position
⌘S	⌃S	preservation
↓	↓	Move cursor down
↑	↑	Move the cursor up

4. View and edit shortcuts

① Check out the shortcuts

① Get into Jupyter Notebook main interface “File” in .

② stay “New” Select the environment from the drop-down list to create a notebook .

③ Click on “Help”.

④ Click on “Keyboard Shortcuts”.

② Edit shortcut

⑴ Method 1

① Get into Jupyter Notebook main interface “File” in .

② stay “New” Select the environment from the drop-down list to create a notebook .

③ Click on “Help”.

④ Click on “Keyboard Shortcuts”.

⑤ In the pop-up dialog box “Command Mode (press Esc to enable)” Side Click “Edit Shortcuts” Button .

⑵ Method 2

① Get into Jupyter Notebook main interface “File” in .

② stay “New” Select the environment from the drop-down list to create a notebook .

③ Click on “Help”.

④ Click on “Edit Keyboard Shortcuts”.

③ example

View and edit shortcuts

8、 ... and 、 Close and exit

1. Turn off the laptop and terminal

When we're in Jupyter Notebook When a terminal or notebook is created in , A new window will pop up to run the terminal or notebook . When we want to exit the terminal or notebook after using it , only 「 Close page 」 You can't end the program , Therefore, we need to turn it off completely through the following steps .

① Method 1

⑴ Get into “Files” page .

⑵ Check the... You want to close “ipynb” The notebook . The icon of the running notebook is green , And the back is marked with “Running” The words... ; The icon of the closed notebook is gray .

⑶ Click the yellow... At the top “Shutdown” Button .

⑷ Successfully closed the notebook .

• Be careful ： This method can only close the notebook , Unable to close the terminal .

② Method 2

⑴ Get into “Running” page .

⑵ The first column is “Terminals”, That is, all running terminals will display ; The second column is “Notebooks”, That is, all running “ipynb” Notebooks will be displayed here .

⑶ Click the terminal or notebook you want to close, and then click yellow “Shutdown” Button .

⑷ Successfully shut down the terminal or notebook .

• Be careful ： This method can close any running terminal and notebook .

③ Be careful

⑴ Only “ipynb” Notebooks and terminals need the above methods to end their operation .

⑵ “txt” file , namely “New” In the drop-down list “Text File”, as well as “Folder” Just close the page where the program is running, and the operation will end , It is not necessary to close... Through the above steps .

④ demonstration

Turn off the laptop and terminal

2. sign out Jupyter Notebook Program

If you want to quit Jupyter Notebook Program , You can't exit just by closing the web page , Because when you open Jupyter Notebook when , It's actually the server that started it .

You can try closing the page , And open a new browser page , Enter the previous address into the address bar , Then jump to the page , You will find that you have entered the just now “ close ” Of Jupyter Notebook page .

If you forget the address of the page you just closed , Can be started at Jupyter Notebook Find the address in your terminal , Copy and paste into the address bar of the new browser page , You will find that you can also enter the page you just closed .

therefore , Want to quit completely Jupyter Notebook, Need to shut down its server . Just press... On the terminal it starts ：

• Mac user ：control c

• Windows user ：ctrl c

Then it will prompt on the terminal ：“Shutdown this notebook server (y/[n])?” Input y You can turn off the server , This is a complete exit Jupyter Notebook Program . here , If you want to access by entering the URL of the page you just closed Jupyter Notebook You will see the error report page .

 Recommended reading :
 introduction :  The most complete zero Foundation Python The problem of   |  Zero Basics 8 Months Python  |  Actual project  | learn Python That's the shortcut 
 dried food : A short comment on crawling Douban , The movie 《 The rest of us 》 | 38 year NBA Best player analysis  |    From people's expectation to public praise ！ Tang Dynasty detective 3 disappointing   |  Laugh at the story of the new Yitian dragon slaying  |  Riddle answer King  ｜ use Python Make a massive sketch of my little sister  ｜ Mission impossible is so hot , I use machine learning to make a mini recommendation system movie 
 Interest : Pinball game   |  squared paper for practicing calligraphy   |  Beautiful flowers  |  Two hundred lines Python《 Cool run every day 》 game !
AI:  A robot that can write poetry  |  Color the picture  |  Forecast revenue  |  Mission impossible is so hot , I use machine learning to make a mini recommendation system movie 
 Gadget : Pdf turn Word, Easily handle forms and watermarks ！ |  One touch html Save the page as pdf！|   bye PDF Withdrawal charges ！ |  use 90 Lines of code create the strongest PDF converter ,word、PPT、excel、markdown、html One click conversion  |  Make a nail low-cost ticket reminder ！ ｜60 Line of code to do a voice wallpaper switcher, look at my little sister every day ！｜

Annual hot money copy

• 1). Oh my god ！Pdf turn Word use Python Easy to handle ！

• 2). learn Python It's delicious ！ I use 100 Line of code to make a website , Help people PS Travel pictures , Earn a chicken leg to eat

• 3). Premiere billions , Hot all over the net , I analyzed 《 My sister 》, Discovered the secrets  

• 4).80 Line code ！ use Python Make a dorai A Dream separation  

• 5). What you have to master 20 individual python Code , short , Useful  

• 6).30 individual Python Strange sexual skills Collection  

• 7). I summed up 80 page 《 Rookie Science Python Select dry goods .pdf》, Is dry  

• 8). bye Python！ I have to learn Go 了 ！2500 Word depth analysis ！

• 9). Found a licking dog welfare ！ This Python Reptile artifact is great , Automatically download sister pictures

Click to read the original , see B My station 20 A video ！