In this part, you need to apply a simple ANN model $($same as in part $1)$ on the

$$MobilePriceDataset$,$ which has $20$ features and one label $$price$\_$range$($as category range

between $0$ to $3).$ Your task is:

$(1)$Simply use the same steps in the previous part to implement ANN.

$(2)$Note that:

a$.$ After reading the dataset you need to change dataframe to numpy by using this code:

#Changing pandas dataframe to numpy array

X$\_$data $=$ data.iloc$[$:$,$:$20].$values

y $=$ data.iloc$[$:$,20$:$21].$values

b$.$ You need to split the dataset into $80\%$ train set and $20\%$ test set.

c$.$ To normalize the data, you can use this method:

#Normalizing the data

from sklearn.preprocessing import StandardScaler

sc $=$ StandardScaler$()$

X$\_$train $=$ sc$.$fit$\_$transform$($X$\_$train$)$

X$\_$test $=$ sc$.$fit$\_$transform$($X$\_$test$)$

d$.$ No need to flatten the data because it is already in one dimensional shape.

e$.$ Define the ANN model with this setting:

model $=$ keras.Sequential$()$

model.add$($keras$.$layers.Dense$(16,$ input$\_$shape$=(20,),$ activation$=$'sigmoid'$))$

model.add$($keras$.$layers.Dense$(12,$ activation$=$'sigmoid'$))$

model.add$($keras$.$layers.Dense$(4,$ activation$=$'softmax'$))$

f$.$ When you train the model change the number of epochs to be any number of your

choice between $100$ and $120.$