Question

In Python. Create a function called predictKNN(). Your function will return the classification of your data-pointIn addition to any parameters you see fit, your function should accept:

• k
• a data-point: a vector of r numbers
• a dataframe with r columns.
  Run your function at least 5 times with different parameters.

Answer 1

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import classification_report, confusion_matrix


# In os.chdir() provide the address of your working directory which is where your dataset is stored.
import os
os.chdir(r"C:\Users\LENOVO\Documents\Data Science\DataSets") 


def predictKNN(k,dataframe):
    #Loading the dataframe
    dataframe = pd.read_csv(dataframe)
    
    # Dividing dataframe into indepent variable(X) and dependent or target variable(Y)
    X = dataframe.iloc[:, :-1].values
    y = dataframe.iloc[:, -1].values
    
    # Dividing dataset into training and testing sets
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20)
    
    # Preprocessing the data
    scaler = StandardScaler()
    scaler.fit(X_train)
    X_train = scaler.transform(X_train)
    X_test = scaler.transform(X_test)
    
    #Applying KNN
    classifier = KNeighborsClassifier(n_neighbors=2)
    classifier.fit(X_train, y_train)
    y_pred = classifier.predict(X_test)
    
    # Checking the performance
    from sklearn.metrics import classification_report, confusion_matrix
    print(confusion_matrix(y_test, y_pred))
    print(classification_report(y_test, y_pred))
    
    return 

Note :

• In order to run successfully this code, make sure you provide right working directory and import right library functions.
• This code is written on assumptions that target variable column is the last column of the dataframe, incase you want another column to be your target variable then you have to manually edit X and Y.
• Since every dataset is unique and different you may need to do some changes in this code in order to successfully run your code, but don't worry this is the main code. You might need to do some data pre processing like converting string columns into integers one or dropping them out which again depends on your dataset and one's personal choice. If you want you can do all data preprocessing step outside the function and then pass X_train, Y_train, y_test, y_train to the function.

I am attaching the code where I used diabetes dataset saved in my pc