Q$1.$

import numpy as np

import pandas as pd

from sklearn.datasets import fetch$\_$openml

from sklearn.preprocessing import StandardScaler

from sklearn.impute import KNNImputer

from sklearn.neighbors import LocalOutlierFactor

Load the diabetes dataset

diabetes $=$ fetch$\_$openml$(\text{'}$diabetes$\text{'},$ version$=1,$ as$\_$frame$=$True$)$

diabetes$\_$df $=$ diabetes.frame

Create a descriptive statistics for the dataset

print$($diabetes$\_$df$.$describe$())$

Normalize the columns 'age' and 'insu'

scaler $=$ StandardScaler$()$

diabetes$\_$df$[$columns$\_$to$\_$normalize$]=$ scaler.fit$\_$transform$($diabetes$\_$df$[$columns$\_$to$\_$normalize$])$

Detect outliers in two columns 'age' and 'insu' using LOF or Isolation forest

diabetes$\_$df$[\text{'}$outlier$\text{'}]=$ lof.fit$\_$predict$($diabetes$\_$df$[$columns$\_$for$\_$outlier$\_$detection$])$

Filter out rows identified as outliers $($rows with outliers will be $-1)$

filtered$\_$df $=$ diabetes$\_$df$[$diabetes$\_$df$[\text{'}$outlier$\text{'}]!=-1]$

Print the length of filtered dataset with and without outliers

print$($f$"$Length of dataset with outliers: $\{$len$($diabetes$\_$df$)\}")$

print$($f$"$Length of dataset without outliers: $\{$len$($filtered$\_$df$)\}")$

Introduce missing values in the 'mass' and 'insu' columns and print the missing indices in those columns

np$.$random.seed$(42)$

diabetes$\_$df$[\text{'}$mass$\_$with$\_$missing'$]=$ diabetes$\_$df$[\text{'}$mass$\text{'}].$copy$()$

diabetes$\_$df$[\text{'}$mass$\_$with$\_$missing'$].$loc$[$np$.$random.choice$($diabetes$\_$df$.$index, size$=20)]=$ np$.$nan

diabetes$\_$df$[\text{'}$insu$\_$with$\_$missing'$]$ diabetes$\_$df$[\text{'}$insu$\text{'}].$copy$()=$