From re-environment construction of Python to graph drawing (on visual studio code)

I want to study machine learning, Because I found a gap time and sporadically moved it by trial and error such as anaconda and vscode The execution environment has begun to get confused. .. .. Since we have rebuilt from zero base, we will summarize the procedure.

This article is composed of these two parts.

• Step 1: Create an environment where python can run
• Step (2): Building an environment for drawing graphs

Specifically, we will do the following.

• Python-Install
• numpy-Import number calculation library
• matplotlib-Import graph drawing library
• Combine the above and try to draw a graph on VScode

Step 1: Create an environment where python can run

Premise

• windows10 home edition
• Built with vscode (not using anaconda)

What to do on Windows

Install python from the following https://www.python.org/downloads/windows/

Choose from Stable Releases This time, I chose "Windows x86-64 executable installer".

The exe file will be dropped, so install it quickly.

What to do on vscode

Then, I would like to recreate the environment with vscode. An icon that looks like the "rice field" on the far left is broken (select the extension)

Then type "python" and install the python that probably hits the top

↓↓

At the same time, the dependent jupyter will also be installed

Then specify the Python path. Select Settings for Preferenes

And if you hit the following command

python.pythonpath

The screen for specifying the path will be displayed, so enter the correct path.

• In my environment, I used anaconda about a year ago, so anaconda was specified by default. ..

To find out the path in windows, I think you should open the executable file.

If you install it normally, you will lose such a path

user

C:\Users\*****\AppData\Roaming\Microsoft\Windows\Start Menu\Programs\Python 3.9

workspace

C:\Users\*****\AppData\Local\Programs\Python\Python39\python.exe

Then, create a hello.py file and execute it with the following contents.

msg = "hello world!"
print(msg)

Execution result ↓

hello world!

Congratulations! Now you have a minimum environment for python to work! !! !!

Step (2): Try drawing a graph

Build the environment

Because numpy and matplotlib are required Install using pip.

• Pip is like nmp called node and gem called ruby.

First of all, let's update pip.

• Probably, it was installed at the same time when python was first installed, Keep the version up to date.

py -m pip install -U pip

To install NumPy, execute the following command.

py -m pip install numpy

To install matplotlib, execute the following command.

py -m pip install matplotlib

The library environment for drawing graphs is now ready.

Actually draw a graph

Let's overwrite and execute hello.py as follows without detailed explanation.

import numpy as np
import matplotlib.pyplot as plt

msg = "Hello numpy and matplotlib"
print(msg)

x = np.linspace(0, 1, 5)
y = x ** 2
plt.plot(x, y)
plt.show()

If you can draw the graph below, you are successful!

The purpose of this article has been achieved, so it ends here! Thank you for your hard work!

bonus

For the purpose of using python If you want to do regression analysis etc., it is a good idea to include the pandas library at the same time.

py -m pip install pandas

With this, you can make up for missing data, sort data, etc. It is a library that is useful when preparing for analysis.