[PYTHON] About Partial Derivatives-Why Solve-

Purpose of partial differentiation

A: To help you understand the possibility of data association and the current situation when there are multiple explanatory variables.

What were you doing in differentiation?

When considering a two-dimensional graph, make a slope in the binary range.

Why ask?

(1) To think more closely about the laws of straight lines and curves at a certain point and make them highly accurate. (2) By doing this, it becomes possible to further grasp the relationship between the two values and apply it to others as a law.

procedure

① Draw a tangent to the first value of the binary. ② Calculate this with a mathematical formula (extreme value)

Next, consider partial differentiation Partial differentiation is performed for the purpose of differentiation in multivariable functions.

2-variable function, 3-variable function, multivariable function

How to do it?

Consider a graph (many inputs, 1 output) in a multivariable function. You can basically think in three dimensions

Fix one point and find the slope (extreme value) from multiple axes.

The graph has an image that is represented in multiple dimensions and becomes a curve

?? ?? Round dee comes out

This symbol is a symbol for distinguishing between partial differential and differential

?? Delta comes out

To represent fine movements in a direction

Finally, I thought about it from the perspective of data analysis

How to utilize this multivariable function? Calculation that appears when the explanatory variables are two or more values in data analysis using machine learning.

merit;

(1) It is possible to determine whether the data is linear or non-linear from the graph.

(2) To find a highly accurate law from the data between the numerical values and improve the accuracy of prediction.

That is, well "Let's differentiate for the time being" Let's find a highly accurate law in the given data! !! It turned out that it was to say.

It may have been easier than I expected.