Problem $1($Coding$)8$ pts: Consider the car evaluation dataset from UCI ML repository for this task. There

are four classes in the dataset. Convert this multiclass classification into binary classification: acceptable

$($acceptable$,$ good, and very good$)$ vs$.$ non$-$acceptable.

a$.(6$ pts$)$ Split the dataset into training and test sets of $80$:$20$ ratio $($use random$\_$seed $=2024)$ and

test$\_$size $=0\mathrm{.}20.$ You must train a Na$$ve Bayes classifier from scratch without using any

package$/$library$.$

b$.(2$ pts$)$ Mention the training and test performance in terms of accuracy. Mention, is there

overfitting issues.

Write code in python.