Load Fisher’s iris data set in Python from sklearn datasets using the command iris = datasets. load_iris(), then taking the transpose of the iris. data array. We will refer to the 4 x 150 dataset as irisInputs. 1. (10 points) Using a 2 x 2 subplot, plot a histogram for each row of irisInputs. Which dimension (row) has the greatest variance? 2. (5 points) Find and report (print) the mean of the dataset. This should be a 1x4 row with a mean value for each dimension in the dataset. 3. (10 points) Create a zero-meaned data matrix Z. You can do this by subtracting the mean column from every column of irisInputs. Hint: you can create a matrix that replicates the mean n times using np. tile(). Report your code only (do not output Z or the mean matrix). 4. (5 points) Calculate the covariance matrix of Z (You can do this by multiplying every element of Z by every other element of Z, then dividing by the number of samples – 1). C = np. matmul(Z, Z.T)/(n-1) Report the size of your covariance matrix C. Is the size correct?