Write a Python function that takes as inputs a vector of values and returns a vector of semivariances, which we will define as the Euclidean distance between each pair of observations: = (x-x), where x and x represent different values. Your code should compute and return the semivariance as a vector, with no repetitions (ie., you can't just use a simple double for-loop because (x-x))= (x;- x)). There are several possible ways to solve the problem, including a modified double for-loop, or write a separate function. def raw vario(y): # Your code here v=None return v Your function should properly handle NaNs in the input (semivariance of NaN) and should raise a ValueError if the input is not a vector. Example inputs and outputs: x=[1, 2, 3], v 10.0, 0.5, 2.0, 0.0, 0.5, 0.01 x=[-2, -1, 0, 1, 2], v = [0.0, 0.5, 2.0, 4.5, 8.0, 0.0, 0.5, 2.0, 4.5, 0.0, 0.5, 2.0, 0.0, 0.5,0.0] x = np.array([[1, 2], [2,2]]), an error: "ValueError: x must be a vector"