Introduction

Here, I will explain the procedure for those who have windows 10 and want to move AI programming ** "quickly move and get an image" **.

It is for ** beginners , the purpose is to prepare the development environment, execute the program (I do not program. Copy and paste is OK ^^;), ** "Oh, AI programming is like this ~ The place called "" is ** GOAL **.

What to do

First, let AI learn ** 60,000 pieces of teacher data, which is a set of `handwritten numbers 0-9 and their correct labels (0-9).

After that, let AI verify ** the 10,000 verification data that is a set of handwritten numbers 0-9 and their correct answer labels (0-9), and the recognition rate and handwritten numbers of the verification data will be determined by AI. Try which number from 0 to 9 you recognized.

`(1) Teacher data (60,000 sheets)`

Have the handwritten image memorized as a number from 0 to 9 along the correct label.

`(2) Verification data (10,000 sheets)`

The handwritten image is classified into numbers from 0 to 9 (classification) by AI, and it is judged whether or not it hits by comparing with the correct answer data.

The above data is automatically downloaded from the MNIST database when the program is executed, so collect image files andcorrect answer labeling (image data and correct answers) (Label binding)etc. ** No preparation is required **.

Development environment installation

First, install what you need for development. The following two software are required for the development environment.

No.	Development environment	software	The version used this time	Description
1	python language	Anaconda	Anaconda Navigator 1.9.12	Anaconda is a free, open source distribution of the Python language aimed at simplifying package management and deployment. It's very convenient.
2	IDE(Integrated development environment)	pycharm	Community 2020.2nd edition	Code analysis and graphical debugger are possible. Community Edition is free.

By the way, we use an open source neural network library called Keras to run AI.

Work flow

Please execute in the order of No. 1 → 4 in the table below.

No.	Work outline	Work content	Description
1	Anaconda settings	①Download ②Install ③ App settings	Enables the use of the Python language to run an open source neural network library called Keras. Python version can call library 3.7.Prepare X.
2	Pycharm settings	①Download ②Install ③ App settings	To make it easier to develop in the python language, we will make an integrated development environment available.
3	AI programming with Python	① AI programming ② Add library ③ AI programming execution	rest assured. I don't program. The program can be copied and pasted.
4	Achievement of purpose	Indulge in thought	Thank you for your hard work!