jupyter what
Jupyter Notebooks is an interactive notebook, run support more than 40 programming languages, its essence is an open source Web applications, we can use it to create and share code and documentation, he can support real-time code, mathematics equation, visualization and markdown.
Jupyter Notebooks provides an environment where you can without leaving the environment in which you can write your code, run the code, look at the output, data visualization and view the results. So this is a handy tool can perform end to end workflow scientific data, including data cleaning, statistical modeling, building, and training machine learning models, data visualization, and so on, it can provide a powerful interactive capabilities for the data scientist and a tool that allows you to run multiple languages, they are more interactive than a simple IDE platform, so they are widely used in a more pedagogical way to show the code
when you're still in the prototype phase, Jupyter Notebooks advantage It becomes more apparent, because your code is written in the form of an independent unit, and these units are individually. This allows the user to test a particular block of code in a project without having to execute code from the beginning of the project. Many other IDE environments (such as RStudio) there are several other ways could do it, but I personally feel that a single cell structure Jupyter is the best.
jupyter installation
jupyter need to support python environment, so you need to install python, then install jupyter, but for new users, you can also use Anaconda to install the release Jupyter Notebooks,
because Anaconda will install this python and Jupyter Both tools, and also contains significant amounts of data science and machine learning communities commonly used software packages. I offer Anaconda installation package download, you can download: https://www.anaconda.com/downloadForms, but this article only describes the installation of a second pip command jupyter.
Jupyter installation using the command form, we first need to install the python environment, we do not describe this step, you can refer https://www.runoob.com/python3/python3-install.html installation;
after installation of python, we pip can use the command to install jupyter, specifically the following step:
(1) check to see whether you need to update the version pip pip, terminal execute:
pip install --upgrade pip
In this step, we may encounter some problems, the following message may appear:
Solution: The problem has been mainly due to the corresponding local version, so when the update will be some conflict, our solution is to find the path to the above file, find the file version conflict in the corresponding file and then delete the , can be performed again;
(2) ipython installation environment, the terminal performs:
pip install ipython
In this step, still there may be conflicts pip above update process, what we need is the will delete all files of conflict, you can execute commands;
(3) installation jupyter, terminal execute:
pip install jupyter
(4) Open jupyter, the terminal performs:
jupyter notebook
By following the above steps, jupyter will be installed, we can open it for use, but in the process that opens, you may experience problems occupied port (if you do not set the amount of time in the open port, it will automatically use 8888), so we need to use the command to set the new port number to:
jupyter notebook --port 9999
Now we can normally use, but we found our stored files will be saved to our jupyter installation path, then we will modify the path to save the file, perform the following steps:
Jupyter notebook --generate-config, where output will show us the path to store configuration files, and asked if we want to modify the file, enter y,: (1) terminal executes
the file contents of our content will overwrite the previous but if we just view the configuration file, be sure to enter n. (2) find c.NotebookApp.notebook = '' attribute, whose front is removed commented code, modify the project storage address in the address set itself; (3) save and exit
Always need to use the terminal command to open jupyter, it seems a bit complicated Oh, let's add a shortcut:
(1) New jupyter.txt file
(2) Find your jupyter installation address: the terminal performs: the WHERE jupyter
(3) Input content, cd modify the path to get to the last step of juoyter address
@echo off D: cd C:\Users\10553\AppData\Local\Programs\Python\Python37\Scripts jupyter notebook
(4) Save the file, modify the file format .bat
this time, we are ready to use when, simply double-click the desktop shortcut to be all right, do not have to use the cumbersome command line. Well, the basic configuration to get, you can open using jupyter friends.