K-Nearest Neighbours

#1

I can’t understand where to begin in the KNN Challenge (Diabetes Classification) as in how to approach a challenge.

#2

Hello Varun,
It’s nothing different that what we do in a normal KNN algorithm. We’ll simply pass the X_train, Y_Train and a testing point (iterating over the X_Test and passing each testing point at once) in the prediction function which we wrote in our KNN Algorithm. Store the returned prediction in a list, then convert it into a CSV file and upload it to get your score.

I hope this helps. If you have a doubt in the concept, please let me know.

#3

How to start with any data science challenge ?

#4

Okay. It seems simple explained in that way. Thank you.

#5

But then what does this mean : “Plot a bar graph showing number of classes and no of examples in each class.” Classes of what?

#6

Classes refers to our labels, i.e., the unique elements in our y_train.