Write a python code with numpy

We can normalise this data further. Whiten the dataset so that it has zero mean and unit covariance. This transforms our dataset so that it is centered on the origin (demeaning) and "spherical" (whitenening with covariance matrix). This is particularly useful if we are going to try and map from data in one vector space to another; having the data in standard scaling, with no offset and no correlation among dimensions makes the data easier to work with. Task B.5 Use the SVD to compute the inverse square root of the covariance matrix whisky_cov. Multiply the demeaned whisky matrix by this to produce whitened_whisky. This represents the data with the mean removed and all correlations eliminated. We can normalise this data further. Whiten the dataset so that it has zero mean and unit covariance. This transforms our dataset so that it is centered on the origin (demeaning) and "spherical" (whitenening with covariance matrix). This is particularly useful if we are going to try and map from data in one vector space to another; having the data in standard scaling, with no offset and no correlation among dimensions makes the data easier to work with. Task B.5 Use the SVD to compute the inverse square root of the covariance matrix whisky_cov. Multiply the demeaned whisky matrix by this to produce whitened_whisky. This represents the data with the mean removed and all correlations eliminated