Machine learning terminology

  1) Model

    The word model will run through the whole tutorial , It is the core concept of machine learning . You can think of it as a “ Magic box ”, You make a wish to it （ input data ）, It will help you realize your wishes （ Output forecast results ）. The whole process of machine learning will revolve around the model , Train the best “ Magic box ”, It can realize your promise as accurately as possible “ desire ”, This is the goal of machine learning .

  2) Data sets

    Data sets , It's easy to understand literally , It represents a collection of data , if “ Model ” yes “ Magic box ” Words , So the data set is responsible for charging it “ Energy battery ”, In short , If a dataset is missing , Then the model has no meaning to exist . The data set can be divided into “ Training set ” and “ Test set ”, They are in the of machine learning “ Training phase ” and “ Prediction output stage ” It plays an important role .

 3) sample & features

    Sample refers to the data in the data set , A piece of data is called “ A sample ”, Usually , The sample will contain multiple eigenvalues to describe the data , For example, there is now a set of data describing human form “180 70 25” If you just look at the data, you will be very confused , But with “ features ” After describing, it will become easy to understand , As shown below ：

height （cm）	weight （kg）	Age
180	70	25

chart 1： sample & features

    It can be seen from the above figure that the composition of the data set is “ One sample per line , A list of characteristics ”. Eigenvalues can also be understood as the correlation of data , The data of each column is related to the eigenvalue of this column .

  

4) vector

    Any algorithm will involve many mathematical terms or formulas . Many mathematical formulas will also be involved in the process of writing this tutorial , And professional terminology , Here, let's briefly explain the common basic terms .

    The first common term is “ vector ”, Vector is the key term of machine learning . Vectors are strictly defined in linear algebra . Vectors are also called Euclidean vectors 、 Geometric vectors 、 vector , A quantity with size and direction . You can visually understand it as a line segment with an arrow . The arrow points to ： Represents the direction of the vector ; segment length ： Represents the size of the vector . The quantity corresponding to a vector is called quantity （ It's called scalar in physics ）, Quantity is only size , No direction . Every sample in the data set is a piece of data in vector form .

  5) matrix

    Matrix is also a commonly used mathematical term , You can think of a matrix as a two-dimensional array of vectors , Data sets store data in the form of two-dimensional matrices , You can think of it as a spreadsheet “ One sample per line , A list of characteristics ” The form of expression is as follows ：