Numpy.Python矩阵乘法

Question

我是python的新手。我试图将图像读入数组并执行SVD（步骤1）。然后将SVD与原始数组相乘（步骤2）。在下面以粗体突出显示的步骤2中，我得到一个错误“形状（10,1024）和（640,1024）未对齐：1024（暗淡1）！= 640（暗淡0）。非常感谢任何帮助。

# Import sample.png into array img as command parameter
      img = scipy.misc.imread(inputfile)
      print(img.shape)

# Perform SVD on im and obtain individual matrices
      P, D, Q = numpy.linalg.svd(img, full_matrices=False)

# Compute overall SVD matrix based on individiual matrices
      svd_decomp = numpy.dot(numpy.dot(P, numpy.diag(D)), Q)

 # Keep Top entries in svd_decomp
      initial = svd_decomp.argsort()
      temp = numpy.array(initial)
      svd_final = numpy.argpartition(temp,-rank)[-rank:]
      print (svd_final.shape)

 # Multiply to obtain the best rank-k approximation of the original array 
 #numpy.dot(svd_decomp * img)
 #print (svd_final.dot(img))
     print(numpy.dot(svd_final,img))**