Using R$,$ we now perform LDA on a subset of the iris dataset which we have already covered in the lecture. The iris data set is a data frame with $150$ samples $($rows$)$ and $5$ variables $($columns$)$ named Sepal.Length, Sepal.Width, Petal.Length, Petal.Width, and Species.

$($a$)(1$ point$)$ Create a subset of the iris data frame, where the rows for the species virginica and the columns Sepal.Length and Petal.Width are excluded.

### Your Solution $($Code$)$

$($b$)(1$ point$)$ Create a scatter plot, where Petal.Length is plotted against Sepal.Width, and where each dot is colored by its corresponding species.

### Your Solution $($Code$)$

$($c$)(1$ point $(0\mathrm{.}5+0\mathrm{.}5)+0\mathrm{.}5$ Bonus points$)$ Perform LDA using the created subset, and print the fitted model. Please explain the values of Prior probabilities of groups, Group means, Coefficients of linear discriminants. $($Hint: The package MASS is required for the lda$()$ function$)$

### Your Solution $($Code$)$

$($d$)(1$ point$)$ Create the plot in $($b$)$ and add the LDA line that discriminates between the two classes.

### Your Solution $($Code$)$

$($e$)(2$ points$)$ Obtain the predicted class using the fitted model on the created training dataset. Further, print the confusion matrix. What is the train accuracy for this classifier? $($help$($predict$))$

### Your Solution $($Code$)$