Create a Python Program, provide codes and detailed explanation at each step.

Goal:

Create a mini but functional kNN classifier with hyperparameter search $($specifically$,$ use GridSearchCV in sklearn.model$\_$selection$).$

Use Numpy for data processing $($data initialisation, data insertion etc$)$ and use Scikit$-$learn library for machine learning part.

The program asks the user for input N $($positive integer$)$ and reads it$.$

Then the program asks the user to provide N $($x$,$ y$)$ pairs $($one by one$)$ and reads all of them: first: x value, then: y value for every pair one by one.

X is treated as the input feature and Y is treated as the class label. X is a real number, Y is a non$-$negative integer.

This set of pairs constitutes the training set TrainS $=\{($x$,$ y$)\_$i$\},$ i $=1..$N$.$

Then the program asks the user for input M $($positive integer$)$ and reads it$.$

Then the program asks the user to provide M $($x$,$ y$)$ pairs $($one by one$)$ and reads all of them: first: x value, then: y value for every pair one by one.

X is treated as the input feature and Y is treated as the class label. X is a real number, Y is a non$-$negative integer.

This set of pairs constitutes the test set TestS $=\{($x$,$ y$)\_$i$\},$ i $=1..$M$.$

In the end, the program outputs: the$$best k$$for the kNN Classification method and the$$corresponding test accuracy.

kNN Classifier should be trained on pairs from TrainS, tested on x values from TestS and compared with y values from TestS.

Try the following range of k: $1<=$ k $<=10.$