how to do Principal Components Analysis of a Matrix

[cfc0712]

I need to do Principal Components (PC) Analysis of a Matrix (X). PC need to be chosen as V(:,end-1,end). May I ask what's mean by "PC is chosen as V(:,end-1,end)" and how to do PC?

I have found eigen value and vector for transpose of X multiple X (XTX), here is my code (using python(x,y) 2.10.7.0):

import numpy as np
X=np.matrix(Xtrain)
print("X")
print(X)
XT=X.transpose()
XTX=np.dot(X,XT)
print('XTX')
print(XTX) 

from numpy import linalg as LA
w,v= LA.eig(XTX)
print('eigen values ')
print(w)
print('eigen vector ')
print(v)

#PCA
from sklearn.decomposition import PCA

def pca2(XTX, pc_count = None):
    return PCA(n_components = 4).fit_transform(XTX)

after running, "ImportError: No module named sklearn.decomposition" is shown. What's the reason? Many thanks for help.